path: root/railties/guides/source/initialization.textile

h2. The Rails Initialization Process

This guide explains the internals of the initialization process in Rails works as of Rails 3.1. It is an extremely in-depth guide and recommended for advanced Rails developers.

* Using +rails server+
* Using Passenger

endprologue.

This guide goes through every single file, class and method call that is required to boot up the Ruby on Rails stack for a default Rails 3.1 application, explaining each part in detail a long the way. For this guide, we will be focusing on how the two most common methods (+rails server+ and Passenger) boot a Rails application.

NOTE: Paths in this guide are relative to Rails or a Rails application unless otherwise specified.

h3. Launch!

As of Rails 3, +script/server+ has become +rails server+. This was done to centralize all rails related commands to one common file.

h4. +bin/rails+

The actual +rails+ command is kept in _bin/rails_ at the and goes like this:

<ruby>
  #!/usr/bin/env ruby

  begin
    require "rails/cli"
  rescue LoadError
    railties_path = File.expand_path('../../railties/lib', __FILE__)
    $:.unshift(railties_path)
    require "rails/cli"
  end
</ruby>

This file will attempt to load +rails/cli+ and if it cannot find it then add the +railties/lib+ path to the load path (+$:+) and will then try to require it again.

h4. +railites/lib/rails/cli.rb+

This file looks like this:

<ruby>
  require 'rbconfig'
  require 'rails/script_rails_loader'

  # If we are inside a Rails application this method performs an exec and thus
  # the rest of this script is not run.
  Rails::ScriptRailsLoader.exec_script_rails!

  require 'rails/ruby_version_check'
  Signal.trap("INT") { puts; exit }

  if ARGV.first == 'plugin'
    ARGV.shift
    require 'rails/commands/plugin_new'
  else
    require 'rails/commands/application'
  end
</ruby>

The +rbconfig+ file here is out of Ruby's standard library and provides us with the +RbConfig+ class which contains useful information dependent on how Ruby was compiled. We'll see this in use in +railties/lib/rails/script_rails_loader+.

<ruby>
require 'pathname'

module Rails
  module ScriptRailsLoader
    RUBY = File.join(*RbConfig::CONFIG.values_at("bindir", "ruby_install_name")) + RbConfig::CONFIG["EXEEXT"]
    SCRIPT_RAILS = File.join('script', 'rails')
    ...

  end
end
</ruby>

The +rails/script_rails_loader+ file uses +RbConfig::Config+ to gather up the +bin_dir+ and +ruby_install_name+ values for the configuration which will result in a path such as +/System/Library/Frameworks/Ruby.framework/Versions/1.8/usr/bin/ruby+, which is the default path on Mac OS X. If you're running Windows the path may be something such as +C:/Ruby192/bin/ruby+. Anyway, the path on your system may be different, but the point of this is that it will point at the known ruby executable location for your install. The +RbConfig::CONFIG["EXEEXT"]+ will suffix this path with ".exe" if the script is running on Windows. This constant is used later on in +exec_script_rails!+. As for the +SCRIPT_RAILS+ console, we'll see that when we get to the +in_rails_application?+ method.

Back in +rails/cli+, the next line is this:

<ruby>
  Rails::ScriptRailsLoader.exec_script_rails!
</ruby>

This method is defined in +rails/script_rails_loader+ like this:

<ruby>
  def self.exec_script_rails!
    cwd = Dir.pwd
    return unless in_rails_application? || in_rails_application_subdirectory?
    exec RUBY, SCRIPT_RAILS, *ARGV if in_rails_application?
    Dir.chdir("..") do
      # Recurse in a chdir block: if the search fails we want to be sure
      # the application is generated in the original working directory.
      exec_script_rails! unless cwd == Dir.pwd
    end
  rescue SystemCallError
    # could not chdir, no problem just return
  end
</ruby>

This method will first check if the current working directory (+cwd+) is a Rails application or is a subdirectory of one. The way to determine this is defined in the +in_rails_application?+ method like this:

<ruby>
  def self.in_rails_application?
    File.exists?(SCRIPT_RAILS)
  end
</ruby>

The +SCRIPT_RAILS+ constant defined earlier is used here, with +File.exists?+ checking for its presence in the current directory. If this method returns +false+, then +in_rails_application_subdirectory?+ will be used:

<ruby>
  def self.in_rails_application_subdirectory?(path = Pathname.new(Dir.pwd))
    File.exists?(File.join(path, SCRIPT_RAILS)) || !path.root? && in_rails_application_subdirectory?(path.parent)
  end
</ruby>

This climbs the directory tree until it reaches a path which contains a +script/rails+ file. If a directory is reached which contains this file then this line will run:

<ruby>
  exec RUBY, SCRIPT_RAILS, *ARGV if in_rails_application?
</ruby>

This is effectively the same as doing +ruby script/rails [arguments]+. Where +[arguments]+ at this point in time is simply "server".

h4. +script/rails+

This file looks like this:

<ruby>
  APP_PATH = File.expand_path('../../config/application',  __FILE__)
  require File.expand_path('../../config/boot',  __FILE__)
  require 'rails/commands'
</ruby>

The +APP_PATH+ constant here will be used later in +rails/commands+. The +config/boot+ file that +script/rails+ references is the +config/boot.rb+ file in our application which is responsible for loading Bundler and setting it up.

h4. +config/boot.rb+

+config/boot.rb+ contains this:

<ruby>
  require 'rubygems'

  # Set up gems listed in the Gemfile.
  gemfile = File.expand_path('../../Gemfile', __FILE__)
  begin
    ENV['BUNDLE_GEMFILE'] = gemfile
    require 'bundler'
    Bundler.setup
  rescue Bundler::GemNotFound => e
    STDERR.puts e.message
    STDERR.puts "Try running `bundle install`."
    exit!
  end if File.exist?(gemfile)
</ruby>

In a standard Rails application, there's a +Gemfile+ which declares all dependencies of the application. +config/boot.rb+ sets +ENV["BUNDLE_GEMFILE"]+ to the location of this file, then requires Bundler and calls +Bundler.setup+ which adds the dependencies of the application (including all the Rails parts) to the load path, making them available for the application to load. The gems that a Rails 3.1 application depends on are as follows:

* abstract (1.0.0)
* actionmailer (3.1.0.beta)
* actionpack (3.1.0.beta)
* activemodel (3.1.0.beta)
* activerecord (3.1.0.beta)
* activeresource (3.1.0.beta)
* activesupport (3.1.0.beta)
* arel (2.0.7)
* builder (3.0.0)
* bundler (1.0.6)
* erubis (2.6.6)
* i18n (0.5.0)
* mail (2.2.12)
* mime-types (1.16)
* polyglot (0.3.1)
* rack (1.2.1)
* rack-cache (0.5.3)
* rack-mount (0.6.13)
* rack-test (0.5.6)
* rails (3.1.0.beta)
* railties (3.1.0.beta)
* rake (0.8.7)
* sqlite3-ruby (1.3.2)
* thor (0.14.6)
* treetop (1.4.9)
* tzinfo (0.3.23)

h4. +rails/commands.rb+

Once +config/boot.rb+ has finished, the next file that is required is +rails/commands+ which will execute a command based on the arguments passed in. In this case, the +ARGV+ array simply contains +server+ which is extracted into the +command+ variable using these lines:

<ruby>
  aliases = {
    "g"  => "generate",
    "c"  => "console",
    "s"  => "server",
    "db" => "dbconsole"
  }

  command = ARGV.shift
  command = aliases[command] || command
</ruby>

If we used <tt>s</tt> rather than +server+, Rails will use the +aliases+ defined in the file and match them to their respective commands. With the +server+ command, Rails will run this code:

<ruby>
  when 'server'
    # Change to the application's path if there is no config.ru file in current dir.
    # This allows us to run script/rails server from other directories, but still get
    # the main config.ru and properly set the tmp directory.
    Dir.chdir(File.expand_path('../../', APP_PATH)) unless File.exists?(File.expand_path("config.ru"))

    require 'rails/commands/server'
    Rails::Server.new.tap { |server|
      # We need to require application after the server sets environment,
      # otherwise the --environment option given to the server won't propagate.
      require APP_PATH
      Dir.chdir(Rails.application.root)
      server.start
    }
</ruby>

This file will change into the root of the directory (a path two directories back from +APP_PATH+ which points at +config/application.rb+), but only if the +config.ru+ file isn't found. This then requires +rails/commands/server+ which requires +action_dispatch+ and sets up the +Rails::Server+ class.

h4. +actionpack/lib/action_dispatch.rb+

Action Dispatch is the routing component of the Rails framework. It depends on Active Support, +actionpack/lib/action_pack.rb+ and +Rack+ being available. The first thing required here is +active_support+.

h4. +active_support/lib/active_support.rb+

This file begins with requiring +active_support/lib/active_support/dependencies/autoload.rb+ which redefines Ruby's +autoload+ method to have a little more extra behaviour especially in regards to eager autoloading. Eager autoloading is the loading of all required classes and  will happen when the +config.cache_classes+ setting is +true+.

The +active_support/lib/active_support/version.rb+ that is also required here simply defines an +ActiveSupport::VERSION+ constant which defines a couple of constants inside this module, the main constant of this is +ActiveSupport::VERSION::STRING+ which returns the current version of ActiveSupport.

The +active_support/lib/active_support.rb+ file simply defines the +ActiveSupport+ module and some autoloads (eager and of the normal variety) for it.

h4. +actionpack/lib/action_dispatch.rb+ cont'd.

Now back to +action_pack/lib/action_dispatch.rb+. The next +require+ in this file is one for +action_pack+, which simply calls +action_pack/version.rb+ which defines +ActionPack::VERSION+ and the constants, much like +ActiveSpport+ does.

After this line, there's a require to +active_model+ which simply defines autoloads for the +ActiveModel+ part of Rails and sets up the +ActiveModel+ module which is used later on.

The last of the requires is to +rack+, which like the +active_model+ and +active_support+ requires before it, sets up the +Rack+ module as well as the autoloads for constants within it.

Finally in +action_dispatch.rb+ the +ActionDispatch+ module and *its* autoloads are declared.

h4. +rails/commands/server.rb+

The +Rails::Server+ class is defined in this file as inheriting from +Rack::Server+. When +Rails::Server.new+ is called, this calls the +initialize+ method in +rails/commands/server.rb+:

<ruby>
  def initialize(*)
    super
    set_environment
  end
</ruby>

Firstly, +super+ is called which calls the +initialize+ method on +Rack::Server+.

h4. Rack: +lib/rack/server.rb+

+Rack::Server+ is responsible for providing a common server interface for all Rack-based applications, which Rails is now a part of.

The +initialize+ method in +Rack::Server+ simply sets a couple of variables:

<ruby>
  def initialize(options = nil)
    @options = options
    @app = options[:app] if options && options[:app]
  end
</ruby>

In this case, +options+ will be +nil+ so nothing happens in this method.

After +super+ has finished in +Rack::Server+, we jump back to +rails/commands/server.rb+. At this point, +set_environment+ is called within the context of the +Rails::Server+ object and this method doesn't appear to do much at first glance:

<ruby>
  def set_environment
    ENV["RAILS_ENV"] ||= options[:environment]
  end
</ruby>

In fact, the +options+ method here does quite a lot. This method is defined in +Rack::Server+ like this:

<ruby>
  def options
    @options ||= parse_options(ARGV)
  end
</ruby>

Then +parse_options+ is defined like this:

<ruby>
  def parse_options(args)
    options = default_options

    # Don't evaluate CGI ISINDEX parameters.
    # http://hoohoo.ncsa.uiuc.edu/cgi/cl.html
    args.clear if ENV.include?("REQUEST_METHOD")

    options.merge! opt_parser.parse! args
    options[:config] = ::File.expand_path(options[:config])
    ENV["RACK_ENV"] = options[:environment]
    options
  end
</ruby>

With the +default_options+ set to this:

<ruby>
  def default_options
    {
      :environment => ENV['RACK_ENV'] || "development",
      :pid         => nil,
      :Port        => 9292,
      :Host        => "0.0.0.0",
      :AccessLog   => [],
      :config      => "config.ru"
    }
  end
</ruby>

There is no +REQUEST_METHOD+ key in +ENV+ so we can skip over that line. The next line merges in the options from +opt_parser+ which is defined plainly in +Rack::Server+

<ruby>
  def opt_parser
    Options.new
  end
</ruby>

The class *is* defined in +Rack::Server+, but is overwritten in +Rails::Server+ to take different arguments. Its +parse!+ method begins like this:

<ruby>
  def parse!(args)
    args, options = args.dup, {}

    opt_parser = OptionParser.new do |opts|
      opts.banner = "Usage: rails server [mongrel, thin, etc] [options]"
      opts.on("-p", "--port=port", Integer,
              "Runs Rails on the specified port.", "Default: 3000") { |v| options[:Port] = v }
  ...
</ruby>

This method will set up keys for the +options+ which Rails will then be able to use to determine how its server should run. After +initialize+ has finished, then the +start+ method will launch the server.

h4. +Rails::Server#start+

This method is defined like this:

<ruby>
  def start
    puts "=> Booting #{ActiveSupport::Inflector.demodulize(server)}"
    puts "=> Rails #{Rails.version} application starting in #{Rails.env} on http://#{options[:Host]}:#{options[:Port]}"
    puts "=> Call with -d to detach" unless options[:daemonize]
    trap(:INT) { exit }
    puts "=> Ctrl-C to shutdown server" unless options[:daemonize]

    #Create required tmp directories if not found
    %w(cache pids sessions sockets).each do |dir_to_make|
      FileUtils.mkdir_p(Rails.root.join('tmp', dir_to_make))
    end

    super
  ensure
    # The '-h' option calls exit before @options is set.
    # If we call 'options' with it unset, we get double help banners.
    puts 'Exiting' unless @options && options[:daemonize]
  end
</ruby>

This is where the first output of the Rails initialization happens. This method creates a trap for +INT+ signals, so if you +CTRL+C+ the server, it will exit the process. As we can see from the code here, it will create the +tmp/cache+, +tmp/pids+, +tmp/sessions+ and +tmp/sockets+ directories if they don't already exist prior to calling +super+. The +super+ method will call +Rack::Server.start+ which begins its definition like this:

<ruby>
  def start
    if options[:warn]
      $-w = true
    end

    if includes = options[:include]
      $LOAD_PATH.unshift(*includes)
    end

    if library = options[:require]
      require library
    end

    if options[:debug]
      $DEBUG = true
      require 'pp'
      p options[:server]
      pp wrapped_app
      pp app
    end
</ruby>

In a Rails application, these options are not set at all and therefore aren't used at all. The first line of code that's executed in this method is a call to this method:

<ruby>
  wrapped_app
</ruby>

This method calls another method:

<ruby>
  @wrapped_app ||= build_app app
</ruby>

Then the +app+ method here is defined like so:

<ruby>
  def app
    @app ||= begin
      if !::File.exist? options[:config]
        abort "configuration #{options[:config]} not found"
      end

      app, options = Rack::Builder.parse_file(self.options[:config], opt_parser)
      self.options.merge! options
      app
    end
  end
</ruby>

The +options[:config]+ value defaults to +config.ru+ which contains this:

<ruby>
  # This file is used by Rack-based servers to start the application.

  require ::File.expand_path('../config/environment',  __FILE__)
  run YourApp::Application
</ruby>


The +Rack::Builder.parse_file+ method here takes the content from this +config.ru+ file and parses it using this code:

<ruby>
  app = eval "Rack::Builder.new {( " + cfgfile + "\n )}.to_app",
    TOPLEVEL_BINDING, config
</ruby>

The <ruby>initialize</ruby> method will take the block here and execute it within an instance of +Rack::Builder+. This is where the majority of the initialization process of Rails happens. The chain of events that this simple line sets off will be the focus of a large majority of this guide. The +require+ line for +config/environment.rb+ in +config.ru+ is the first to run:

<ruby>
  require ::File.expand_path('../config/environment',  __FILE__)
</ruby>

h4. +config/environment.rb+

This file is the common file required by +config.ru+ (+rails server+) and Passenger. This is where these two ways to run the server meet; everything before this point has been Rack and Rails setup.

This file begins with requiring +config/application.rb+.

h4. +config/application.rb+

This file requires +config/boot.rb+, but only if it hasn't been required before, which would be the case in +rails server+ but *wouldn't* be the case with Passenger.

Then the fun begins! The next line is:

<ruby>
  require 'rails/all'
</ruby>

h4 +railties/lib/rails/all.rb+


This file is responsible for requiring all the individual parts of Rails like so:

<ruby>
  require "rails"

  %w(
    active_record
    action_controller
    action_mailer
    active_resource
    rails/test_unit
  ).each do |framework|
    begin
      require "#{framework}/railtie"
    rescue LoadError
    end
  end
</ruby>

First off the line is the +rails+ require itself.

h4. +railties/lib/rails.rb+

This file is responsible for the initial definition of the +Rails+ module and, rather than defining the autoloads like +ActiveSupport+, +ActionDispatch+ and so on, it actually defines other functionality. Such as the +root+, +env+ and +application+ methods which are extremely useful in Rails 3 applications.

However, before all that takes place the +rails/ruby_version_check+ file is required first.

h4. +railties/lib/rails/ruby_version_check.rb+

This file simply checks if the Ruby version is less than 1.8.7 or is 1.9.1 and raises an error if that is the case. Rails 3 simply will not run on earlier versions of Ruby than 1.8.7 or 1.9.1.

NOTE: You should always endeavour to run the latest version of Ruby with your Rails applications. The benefits are many, including security fixes and the like, and very often there is a speed increase associated with it. The caveat is that you could have code that potentially breaks on the latest version, which should be fixed to work on the latest version rather than kept around as an excuse not to upgrade.

h4. +active_support/core_ext/kernel/reporting.rb+

This is the first of the many Active Support core extensions that come with Rails. This one in particular defines methods in the +Kernel+ module which is mixed in to the +Object+ class so the methods are available on +main+ and can therefore be called like this:

<ruby>
  silence_warnings do
    # some code
  end
</ruby>

These methods can be used to silence STDERR responses and the +silence_stream+ allows you to also silence other streams. Additionally, this mixin allows you to suppress exceptions and capture streams. For more information see the "Silencing Warnings, Streams, and Exceptions":http://guides.rubyonrails.org/active_support_core_extensions.html#silencing-warnings-streams-and-exceptions section from the Active Support Core Extensions Guide.

h4. +active_support/core_ext/logger.rb+

The next file that is required is another Active Support core extension, this time to the +Logger+ class. This begins by defining the +around_[level]+ helpers for the +Logger+ class as well as other methods such as a +datetime_format+ getter and setter for the +formatter+ object tied to a +Logger+ object.

For more information see the "Extensions to Logger":http://guides.rubyonrails.org/active_support_core_extensions.html#extensions-to-logger section from the Active Support Core Extensions Guide.

h4. +railties/lib/rails/application.rb+

The next file required by +railties/lib/rails.rb+ is +application.rb+. This file defines the +Rails::Application+ constant which the application's class defined in +config/application.rb+ in a standard Rails application depends on. Before the +Rails::Application+ class is defined however, there's some other files that get required first.

The first of these is +active_support/core_ext/hash/reverse_merge+ which can be "read about in the Active Support Core Extensions guide":http://guides.rubyonrails.org/active_support_core_extensions.html#merging under the "Merging" section.

h4. +active_support/file_update_checker.rb+

The +ActiveSupport::FileUpdateChecker+ class defined within this file is responsible for checking if a file has been updated since it was last checked. This is used for monitoring the routes file for changes during development environment runs.






**** REVIEW IS HERE ****

This defines two methods on the module itself by using the familiar +class << self+ syntax. This allows you to call them as if they were class methods: +ActiveSupport.on_load_all+ and +ActiveSupport.load_all!+ respectively. The first method simply adds loading hooks to save them up for loading later on when +load_all!+ is called. By +call+'ing the block, the classes will be loaded. (NOTE: kind of guessing, I feel 55% about this).

The +on_load_all+ method is called later with the +Dependencies+, +Deprecation+, +Gzip+, +MessageVerifier+, +Multibyte+ and +SecureRandom+. What each of these modules do will be covered later.

This file goes on to define some classes that will be automatically loaded using Ruby's +autoload+ method, but not before including Rails's own variant of the +autoload+ method from _active_support/dependencies/autoload.rb_:


<ruby>
  require "active_support/inflector/methods"
  require "active_support/lazy_load_hooks"

  module ActiveSupport
    module Autoload
      def self.extended(base)
        base.extend(LazyLoadHooks)
      end

      @@autoloads = {}
      @@under_path = nil
      @@at_path = nil
      @@eager_autoload = false

      def autoload(const_name, path = @@at_path)
        full = [self.name, @@under_path, const_name.to_s, path].compact.join("::")
        location = path || Inflector.underscore(full)

        if @@eager_autoload
          @@autoloads[const_name] = location
        end
        super const_name, location
      end

      ...
    end
  end
</ruby>

h4. Lazy Hooks

+ActiveSupport::LazyLoadHooks+ is responsible for defining methods used for running hooks that are defined during the initialization process, such as the one defined inside the +active_record.initialize_timezone+ initializer:

<ruby>
  initializer "active_record.initialize_timezone" do
    ActiveSupport.on_load(:active_record) do
      self.time_zone_aware_attributes = true
      self.default_timezone = :utc
    end
  end
</ruby>

When the initializer runs it invokes method +on_load+ for +ActiveRecord+ and the block passed to it would be called only when +run_load_hooks+ is called.
When the entirety of +activerecord/lib/active_record/base.rb+ has been evaluated then +run_load_hooks+ is invoked. The very last line of +activerecord/lib/active_record/base.rb+ is:

<ruby>
ActiveSupport.run_load_hooks(:active_record, ActiveRecord::Base)
</ruby>

h4. +require 'active_support'+ cont'd.

This file also uses the method +eager_autoload+ also defined in _active_support/dependencies/autoload.rb_:

<ruby>
  def eager_autoload
    old_eager, @@eager_autoload = @@eager_autoload, true
    yield
  ensure
    @@eager_autoload = old_eager
  end
</ruby>

As you can see for the duration of the +eager_autoload+ block the class variable +@@eager_autoload+ is set to +true+, which has the consequence of when +autoload+ is called that the location of the file for that specific +autoload+'d constant is added to the +@@autoloads+ hash initialized at the beginning of this module declaration. So now that you have part of the context, here's the other, the code from _activesupport/lib/active_support.rb_:

<ruby>
  require "active_support/dependencies/autoload"

  module ActiveSupport
    extend ActiveSupport::Autoload

    autoload :DescendantsTracker
    autoload :FileUpdateChecker
    autoload :LogSubscriber
    autoload :Notifications

    # TODO: Narrow this list down
    eager_autoload do
      autoload :BacktraceCleaner
      autoload :Base64
      autoload :BasicObject
      autoload :Benchmarkable
      autoload :BufferedLogger
      autoload :Cache
      autoload :Callbacks
      autoload :Concern
      autoload :Configurable
      autoload :Deprecation
      autoload :Gzip
      autoload :Inflector
      autoload :JSON
      autoload :Memoizable
      autoload :MessageEncryptor
      autoload :MessageVerifier
      autoload :Multibyte
      autoload :OptionMerger
      autoload :OrderedHash
      autoload :OrderedOptions
      autoload :Rescuable
      autoload :SecureRandom
      autoload :StringInquirer
      autoload :XmlMini
    end

    autoload :SafeBuffer, "active_support/core_ext/string/output_safety"
    autoload :TestCase
  end

  autoload :I18n, "active_support/i18n"
</ruby>

So we know the ones in +eager_autoload+ are eagerly loaded and it does this by storing them in an +@@autoloads+ hash object and then loading them via +eager_autoload!+ which is called via the +preload_frameworks+ initializer defined in _railties/lib/rails/application/bootstrap.rb_.

The classes and modules that are not +eager_autoload+'d are automatically loaded as they are references

Note: What does it means to be autoloaded? An example of this would be calling the +ActiveSupport::TestCase+ class which hasn't yet been initialized. Because it's been specified as an +autoload+ Ruby will require the file that it's told to. The file it requires is not defined in the +autoload+ call here but, as you may have seen, in the +ActiveSupport::Autoload.autoload+ definition. So once that file has been required Ruby will try again and then if it still can't find it it will throw the all-too-familiar +uninitialized constant+ error.

h4. +require 'action_dispatch'+

Back in _actionpack/lib/action_dispatch.rb_, the next require after _active_support_ is to _active_support/dependencies/autoload_ but this file has already been loaded by _activesupport/lib/active_support.rb_ and so will not be loaded again. The next require is to Rack itself:

<ruby>
  require 'rack'
</ruby>

As mentioned previously, Bundler has added the gems' _lib_ directories to the load path so this _rack_ file that is referenced lives in the Rack gem: _lib/rack.rb_. This loads Rack so we can use it later on when we define +Rails::Server+ to descend from +Rack::Server+.

This file then goes on to define the +ActionDispatch+ module and it's related autoloads:

<ruby>
  module Rack
    autoload :Test, 'rack/test'
  end

  module ActionDispatch
    extend ActiveSupport::Autoload

    autoload_under 'http' do
      autoload :Request
      autoload :Response
    end

    autoload_under 'middleware' do
      autoload :Callbacks
      autoload :Cascade
      autoload :Cookies
      autoload :Flash
      autoload :Head
      autoload :ParamsParser
      autoload :RemoteIp
      autoload :Rescue
      autoload :ShowExceptions
      autoload :Static
    end

    autoload :MiddlewareStack, 'action_dispatch/middleware/stack'
    autoload :Routing

    module Http
      extend ActiveSupport::Autoload

      autoload :Cache
      autoload :Headers
      autoload :MimeNegotiation
      autoload :Parameters
      autoload :FilterParameters
      autoload :Upload
      autoload :UploadedFile, 'action_dispatch/http/upload'
      autoload :URL
    end

    module Session
      autoload :AbstractStore, 'action_dispatch/middleware/session/abstract_store'
      autoload :CookieStore,   'action_dispatch/middleware/session/cookie_store'
      autoload :MemCacheStore, 'action_dispatch/middleware/session/mem_cache_store'
    end

    autoload_under 'testing' do
      autoload :Assertions
      autoload :Integration
      autoload :PerformanceTest
      autoload :TestProcess
      autoload :TestRequest
      autoload :TestResponse
    end
  end

  autoload :Mime, 'action_dispatch/http/mime_type'
</ruby>

h4. +require "rails/commands/server"+

Now that Rails has required Action Dispatch and it has required Rack, Rails can now go about defining the +Rails::Server+ class:

<ruby>
  module Rails
    class Server < ::Rack::Server

      ...

      def initialize(*)
        super
        set_environment
      end

      ...

      def set_environment
        ENV["RAILS_ENV"] ||= options[:environment]
      end
      ...
    end
  end
</ruby>

h4. +require "rails/commands"+

Back in _rails/commands_ Rails calls +Rails::Server.new+ which calls the +initialize+ method on the +Rails::Server+ class, which calls +super+, meaning it's actually calling +Rack::Server#initialize+, with it being defined like this:

<ruby>
  def initialize(options = nil)
    @options = options
  end
</ruby>

The +options+ method like this:

<ruby>
  def options
    @options ||= parse_options(ARGV)
  end
</ruby>

The +parse_options+ method like this:

<ruby>
  def parse_options(args)
    options = default_options

    # Don't evaluate CGI ISINDEX parameters.
    # http://hoohoo.ncsa.uiuc.edu/cgi/cl.html
    args.clear if ENV.include?("REQUEST_METHOD")

    options.merge! opt_parser.parse! args
    options
  end
</ruby>

And +default_options+ like this:

<ruby>
  def default_options
    {
      :environment => "development",
      :pid         => nil,
      :Port        => 9292,
      :Host        => "0.0.0.0",
      :AccessLog   => [],
      :config      => "config.ru"
    }
  end
</ruby>

Here it is important to note that the default environment is _development_. After +Rack::Server#initialize+ has done its thing it returns to +Rails::Server#initialize+ which calls +set_environment+:

<ruby>
  def set_environment
    ENV["RAILS_ENV"] ||= options[:environment]
  end
</ruby>

From the information given we can determine that +ENV["RAILS_ENV"]+ will be set to _development_ if no other environment is specified.

Finally, after +Rails::Server.new+ has executed, there is one more require:

<ruby>
  require APP_PATH
</ruby>

+APP_PATH+ was previously defined as _config/application.rb_ in the application's root, and so that is where Rails will go next.

h4. +require APP_PATH+

This file is _config/application.rb_ in your application and makes two requires to begin with:

<ruby>
  require File.expand_path('../boot', __FILE__)
  require 'rails/all'
</ruby>

The +../boot+ file it references is +config/boot.rb+, which was loaded earlier in the initialization process and so will not be loaded again.

If you generate the application with the +-O+ option this will put a couple of pick-and-choose requirements at the top of your _config/application.rb_ instead:

<ruby>
  # Pick the frameworks you want:
  # require "active_record/railtie"
  require "action_controller/railtie"
  require "action_mailer/railtie"
  require "active_resource/railtie"
  require "rails/test_unit/railtie"
</ruby>

For the purposes of this guide, will will assume only:

<ruby>
  require 'rails/all'
</ruby>

h4. +require "rails/all"+

Now we'll dive into the internals of the pre-initialization stage of Rails. The file that is being required is _railties/lib/rails/all.rb_. The first line in this file is:

<ruby>
  require 'rails'
</ruby>

h4. +require 'rails'+

This file (_railties/lib/rails.rb_) requires the very, very basics that Rails needs to get going. I'm not going to delve into these areas yet, just cover them briefly for now. Later on we will go through the ones that are important to the boot procedure.

<ruby>
  require 'pathname'

  require 'active_support'
  require 'active_support/core_ext/kernel/reporting'
  require 'active_support/core_ext/logger'

  require 'rails/application'
  require 'rails/version'
  require 'rails/deprecation'
  require 'rails/log_subscriber'
  require 'rails/ruby_version_check'

  require 'active_support/railtie'
  require 'action_dispatch/railtie'
</ruby>

+require 'pathname'+ requires the Pathname class which is used for returning a Pathname object for +Rails.root+. Although is coming to use this path name to generate paths as below:

<ruby>
  Rails.root.join("app/controllers")
</ruby>

Pathname can also be converted to string, so the following syntax is preferred:

<ruby>
  "#{Rails.root}/app/controllers"
</ruby>


This works because Ruby automatically handles file path conversions. Although this is not new to Rails 3 (it was available in 2.3.5), it is something worthwhile pointing out.

Inside this file there are other helpful helper methods defined, such as +Rails.root+, +Rails.env+, +Rails.logger+ and +Rails.application+.

The first require:

<ruby>
  require 'active_support'
</ruby>

Is not ran as this was already required by _actionpack/lib/action_dispatch.rb_.


h4. +require 'active_support/core_ext/kernel/reporting'+

This file extends the +Kernel+ module, providing the methods +silence_warnings+, +enable_warnings+, +with_warnings+, +silence_stderr+, +silence_stream+ and +suppress+. The API documentation on these overridden methods is fairly good and if you wish to know more "have a read.":http://api.rubyonrails.org/classes/Kernel.html

For information on this file see the "Core Extensions" guide. TODO: link to guide.

h4. +require 'active_support/core_ext/logger'+

For information on this file see the "Core Extensions" guide. TODO: link to guide.

h4. +require 'rails/application'+

Here's where +Rails::Application+ is defined. This is the superclass of +YourApp::Application+ from _config/application.rb_ and the subclass of +Rails::Engine+ This is the main entry-point into the Rails initialization process as when your application is initialized, your class is the basis of its configuration.

This file requires three important files before +Rails::Application+ is defined: _rails/railties_path.rb_, _rails/plugin.rb_ and _rails/engine.rb_.


h4. +require 'rails/railties_path'+

This file serves one purpose:

<ruby>
  RAILTIES_PATH = File.expand_path(File.join(File.dirname(__FILE__), '..', '..'))
</ruby>

Helpful, hey? One must wonder why they just didn't define it outright.


h4. +require 'rails/plugin'+

Firstly this file requires _rails/engine.rb_, which defines our +Rails::Engine+ class, explained in the very next section.

This file defines a class called +Rails::Plugin+ which descends from +Rails::Engine+.

This defines the first few initializers for the Rails stack:

* load_init_rb
* sanity_check_railties_collisons

These are explained in the Initialization section. TODO: First write initialization section then come back here and link.
TODO: Expand.

h4. +require 'rails/engine'+

This file requires _rails/railtie.rb_ which defines +Rails::Railtie+.

+Rails::Engine+ defines a couple of further initializers for your application:

* set_load_path
* set_autoload_paths
* add_routing_paths
* add_routing_namespaces
* add_locales
* add_view_paths
* add_generator_templates
* load_application_initializers
* load_application_classes

These are explained in the Initialization section. TODO: First write initialization section then come back here and link.

Also in here we see that a couple of methods are +delegate+'d:

<ruby>
  delegate :middleware, :paths, :root, :to => :config
</ruby>

This means when you call either the +middleware+, +paths+ or +root+ methods you are in reality calling +config.middleware+, +config.paths+ and +config.root+ respectively.

+Rails::Engine+ descends from +Rails::Railtie+.

h4. +require 'rails/railtie'+

+Rails::Railtie+ (_pronounced Rail-tie, as in a bowtie_), provides a method of classes to hook into Rails, providing them with methods to add generators, rake tasks and subscribers. All of the facets of Rails are their own Railtie. and as you've probably already figured out, the engines that you use are railties too. Plugins also can be railties, but they do not have to be.

Here there's requires to _rails/initializable.rb_ and and _rails/configurable.rb_.

h4. +require 'rails/initializable'+

The +Rails::Initializable+ module includes methods helpful for the initialization process in rails, such as the method to define initializers: +initializer+. This is included into +Rails::Railtie+ so it's available there as well as +Rails::Engine+, +Rails::Application+ and +YourApp::Application+. In here we also see the class definition for +Rails::Initializer+, the class for all initializer objects.

h4. +require 'rails/configuration'+

The +Rails::Configuration+ module sets up shared configuration for applications, engines and plugins alike.

At the top of this file there are three +require+s:

<ruby>
  require 'active_support/ordered_options'
  require 'rails/paths'
  require 'rails/rack'
</ruby>

h4. +require 'active_support/ordered_options'+

+ActiveSupport::OrderedOptions+ is a special-purpose +OrderedHash+, used for keeping track of the options specified in the configuration of your application.

TODO: expand.

h4. +require 'rails/paths'+

This file is used to set up the +Rails::Paths+ module which is used to set up helpers for referencing paths to the folders of your Rails application, such as in _railties/lib/rails/engine/configuration.rb_ where it is used to firstly define them:

<ruby>
  def paths
    @paths ||= begin
      paths = Rails::Paths::Root.new(@root)
      paths.app                 "app",                 :eager_load => true, :glob => "*"
      paths.app.controllers     "app/controllers",     :eager_load => true
      paths.app.helpers         "app/helpers",         :eager_load => true
      paths.app.models          "app/models",          :eager_load => true
      paths.app.mailers         "app/mailers",         :eager_load => true
      paths.app.views           "app/views",           :eager_load => true
      paths.lib                 "lib",                 :load_path => true
      paths.lib.tasks           "lib/tasks",           :glob => "**/*.rake"
      paths.lib.templates       "lib/templates"
      paths.config              "config"
      paths.config.initializers "config/initializers", :glob => "**/*.rb"
      paths.config.locales      "config/locales",      :glob => "*.{rb,yml}"
      paths.config.routes       "config/routes.rb"
      paths.public              "public"
      paths.public.javascripts  "public/javascripts"
      paths.public.stylesheets  "public/stylesheets"
      paths
    end
  end
</ruby>

You can then get to these helper methods by calling +YourApp::Application.config.paths+.

h4. +require 'rails/rack'+

This file sets up some +autoload+'d constants for +Rails::Rack+:

<ruby>
  module Rails
    module Rack
      autoload :Debugger,  "rails/rack/debugger"
      autoload :Logger,    "rails/rack/logger"
      autoload :LogTailer, "rails/rack/log_tailer"
      autoload :Static,    "rails/rack/static"
    end
  end
</ruby>

h4. +require 'rails/version'+

Now we're back to _rails.rb_. The line after +require 'rails/application'+ in _rails.rb_ is:

<ruby>
  require 'rails/version'
</ruby>

The code in this file declares +Rails::VERSION+ so that the version number can easily be accessed. It stores it in constants, with the final version number being attainable by calling +Rails::VERSION::STRING+.

h4. +require 'rails/deprecation'+

This sets up a couple of familiar constants: +RAILS_ENV+, +RAILS_ROOT+ and +RAILS_DEFAULT_LOGGER+ to still be usable, but raise a deprecation warning when they are. Their alternatives are now +Rails.env+, +Rails.root+ and +Rails.logger+ respectively.

If you wish to know more about how they're deprecated see the +require 'active_support/deprecation/proxy_wrappers'+ section. TODO: link to section.

h4. +require 'rails/log_subscriber'+

The +Rails::LogSubscriber+ provides a central location for logging in Rails 3 so as to not slow down the main thread. When you call one of the logging methods (+info+, +debug+, +warn+, +error+, +fatal+ or +unknown+) from the +Rails::LogSubscriber+ class or one of its subclasses this will notify the Rails logger to log this call in the fashion you specify, but will not write it to the file. The file writing is done at the end of the request, courtesy of the +Rails::Rack::Logger+ middleware.

Each Railtie defines its own class that descends from +Rails::LogSubscriber+ with each defining its own methods for logging individual tasks.

h4. +require 'rails/ruby_version_check'+

This file ensures that you're running a minimum of 1.8.7. If you're running an older version, it will tell you:

<pre>
  Rails requires Ruby version 1.8.7 or later.
  You're running [your Ruby version here]; please upgrade to continue.
</pre>

h4. +require 'activesupport/railtie'+

This file declares two Railties, one for ActiveSupport and the other for I18n. In these Railties there's the following initializers defined:

* active_support.initialize_whiny_nils
* active_support.initialize_time_zone

* i18n.initialize

This Railtie also defines an an +after_initialize+ block, which will (as the name implies) be ran after the initialization process. More on this later. TODO: When you write the section you can link to it.

h4. +require 'action_dispatch/railtie'+

This file is explained in the ActionDispatch Railtie Section. TODO: link

h4. Return to _rails/all.rb_

Now that we've covered the extensive process of what the first line does in this file, lets cover the remainder:

<ruby>
  %w(
    active_record
    action_controller
    action_mailer
    active_resource
    rails/test_unit
  ).each do |framework|
    begin
      require "#{framework}/railtie"
    rescue LoadError
    end
  end
</ruby>

As you may be able to tell from the code, this is going through and loading all the Railties for Active Record, Action Controller, Action Mailer, Active Resource. Two other Railties, one for Active Support and one for Action Dispatch were required earlier, but are still covered in this section for continuity reasons. TODO: link.

h4. ActiveSupport Railtie

From Active Support's README:

Active Support is a collection of various utility classes and standard library extensions that were found useful for Rails.

TODO: Quotify.

h5. +require 'active_support/railtie'+


h4. Active Record Railtie

The Active Record Railtie takes care of hooking Active Record into Rails. This depends on Active Support, Active Model and Arel. From Active Record's readme:

TODO: Quotify.

<plain>
  Active Record connects business objects and database tables to create a persistable domain model where logic and data are presented in one wrapping. It's an implementation of the object-relational mapping (ORM) pattern by the same name as described by Martin Fowler:

    "An object that wraps a row in a database table or view, encapsulates
         the database access, and adds domain logic on that data."

  Active Record's main contribution to the pattern is to relieve the original of two stunting problems:
  lack of associations and inheritance. By adding a simple domain language-like set of macros to describe
  the former and integrating the Single Table Inheritance pattern for the latter, Active Record narrows the
  gap of functionality between the data mapper and active record approach.
</plain>

h5. +require "active_record/railtie"+

The _activerecord/lib/active_record/railtie.rb_ file defines the Railtie for Active Record.

This file first requires Active Record, the _railties/lib/rails.rb_ file which has already been required and so will be ignored, and the Active Model Railtie:

<ruby>
  require "active_record"
  require "rails"
  require "active_model/railtie"
</ruby>

Active Model's Railtie is covered in the next section. TODO: Section.

h5. +require "active_record"+

TODO: Why are +activesupport_path+ and +activemodel_path+ defined here?

The first three requires require ActiveSupport, Active Model and Arel in that order:

<ruby>
  require 'active_support'
  require 'active_model'
  require 'arel'
</ruby>


h5. +require "active_support"+

This was loaded earlier by _railties/lib/rails.rb_. This line is here as a safeguard for when Active Record is loaded outside the scope of Rails.

h5. +require "active_model"+

TODO: Again with the +activesupport_path+!

Here we see another +require "active_support"+ this is again, a safeguard for when Active Model is loaded outside the scope of Rails.

This file defines a few +autoload+'d modules for Active Model, requires +active_support/i18n+ and adds the default translation file for Active Model to +I18n.load_path+.

The +require 'active_support/i18n'+ just loads I18n and adds Active Support's default translations file to +I18n.load_path+ too:

<ruby>
  require 'i18n'
  I18n.load_path << "#{File.dirname(__FILE__)}/locale/en.yml
</ruby>


h5. +require "arel"+

This file in _arel/lib/arel.rb_ loads a couple of Active Support things first:

<ruby>
  require 'active_support/inflector'
  require 'active_support/core_ext/module/delegation'
  require 'active_support/core_ext/class/attribute_accessors'
</ruby>

These files are explained in the "Common Includes" section.

h5. +require 'arel'+

Back in _arel/lib/arel.rb_, the next two lines require Active Record parts:

<ruby>
  require 'active_record'
  require 'active_record/connection_adapters/abstract/quoting'
</ruby>

Because we're currently loading _active_record.rb_, it skips right over it.

h5. +require 'active_record/connection_adapters/abstract/quoting'+

_activerecord/lib/active_record/connection_adapters/abstract/quoting.rb_ defines methods used for quoting fields and table names in Active Record.

TODO: Explain why this is loaded especially.

h5. +require 'active_record'+

Back the initial require from the _railtie.rb_.

The _active_support_ and _active_model_ requires are again just an insurance for if we're loading Active Record outside of the scope of Rails. In _active_record.rb_ the ActiveRecord +Module+ is initialized and in it there is defined a couple of +autoloads+ and +eager_autoloads+.

There's a new method here called +autoload_under+ which is defined in +ActiveSupport::Autoload+. This sets the autoload path to temporarily be the specified path, in this case +relation+ for the +autoload+'d classes inside the block.

Inside this file the +AttributeMethods+, +Locking+ and +ConnectionAdapter+ modules are defined inside the +ActiveRecord+ module. The second to last line tells Arel what SQL engine we want to use. In this case it's +ActiveRecord::Base+. The final line adds in the translations for Active Record which are only for if a record is invalid or non-unique.

h5. +require 'rails'+

As mentioned previously this is skipped over as it has been already loaded. If you'd still like to see what this file does go to section TODO: section.

h5. +require 'active_model/railtie'+

This is covered in the Active Model Railtie section. TODO: link there.

h5. +require 'action_controller/railtie'+

This is covered in the Action Controller Railtie section. TODO: link there.

h5. The Active Record Railtie

Inside the Active Record Railtie the +ActiveRecord::Railtie+ class is defined:

<ruby>
  module ActiveRecord
    class Railtie < Rails::Railtie

    ...
    end
  end
</ruby>

TODO: Explain the logger.

By doing this the +ActiveRecord::Railtie+ class gains access to the methods contained within +Rails::Railtie+ such as +rake_tasks+, +log_subscriber+ and +initiailizer+, all of which the Railtie is using in this case. The initializers defined here are:

* active_record.initialize_timezone
* active_record.logger
* active_record.set_configs
* active_record.initialize_database
* active_record.log_runtime
* active_record.initialize_database_middleware
* active_record.load_observers
* active_record.set_dispatch_hooks

As with the engine initializers, these are explained later.


h4. Active Model Railtie

This Railtie is +require+'d by Active Record's Railtie.

From the Active Model readme:

<plain>
  Prior to Rails 3.0, if a plugin or gem developer wanted to be able to have an object interact with Action Pack helpers, it was required to either copy chunks of code from Rails, or monkey patch entire helpers to make them handle objects that did not look like Active Record.  This generated code duplication and fragile applications that broke on upgrades.

  Active Model is a solution for this problem.

  Active Model provides a known set of interfaces that your objects can implement to then present a common interface to the Action Pack helpers.
</plain>


h5. +require "active_model/railtie"+

This Railtie file, _activemodel/lib/active_model/railtie.rb_ is quite small and only requires in +active_model+. As mentioned previously, the require to _rails_ is skipped over as it has been already loaded. If you'd still like to see what this file does go to section TODO: section.

<ruby>
  require "active_model"
  require "rails"
</ruby>

h5. +require "active_model"+

Active Model depends on Active Support and ensures it is required by making a +require 'active_support'+ call. It has already been loaded from _railties/lib/rails.rb_ so will not be reloaded for us here. The file goes on to define the +ActiveModel+ module and all of its autoloaded classes. This file also defines the english translations for some of the validation messages provided by Active Model, such as "is not included in the list" and "is reserved".

h4. Action Controller Railtie

The Action Controller Railtie takes care of all the behind-the-scenes code for your controllers; it puts the C into MVC; and does so by implementing the +ActionController::Base+ class which you may recall is where your +ApplicationController+ class descends from.

h5. +require 'action_controller/railtie'+

This first makes a couple of requires:

<ruby>
  require "action_controller"
  require "rails"
  require "action_view/railtie"
</ruby>

The _action_controller_ file is explained in the very next section. The require to _rails_ is requiring the already-required _railties/lib/rails.rb_. If you wish to know about the require to _action_view/railtie_ this is explained in the Action View Railtie section.

h5. +require 'action_controller+

This file, _actionpack/lib/action_controller.rb_, defines the Action Controller module and its relative autoloads. Before it does any of that it makes two requires: one to _abstract_controller_, explored next, and the other to _action_dispatch_, explored directly after that.

h5. +require 'abstract_controller'+

+AbstractController+ provides the functionality of TODO.

This file is in _actionpack/lib/abstract_controller.rb_ and begins by attempting to add the path to Active Support to the load path, which it would succeed in if it wasn't already set by anything loaded before it. In this case, it's not going to be set due to Arel already loading it in (TODO: right?).

The next thing in this file four +require+ calls:

<ruby>
  require 'active_support/ruby/shim'
  require 'active_support/dependencies/autoload'
  require 'active_support/core_ext/module/attr_internal'
  require 'active_support/core_ext/module/delegation'
</ruby>

After these require calls the +AbstractController+ module is defined with some standard +autoload+'d classes.


h5. +require 'active_support/ruby/shim'+

This file is explained in the "Common Includes" section beneath.

h5. +require 'active_support/dependencies/autoload+

This file was loaded upon the first require of +active_support+ and is not included. If you wish to be refreshed on what this file performs visit TODO: link to section.

h5. +require 'active_support/core_ext/module/attr_internal'+

This file is explained in the "Common Includes" section as it is required again later on. See the TODO: section. I also think this may be explained in the Active Support Core Extensions guide.

h5. +require 'active_support/core_ext/module/delegation'+

This file is explained in the "Common Includes" section as it has already been required by Arel at this point in the initialization process (see: section TODO: LINK!).

h5. +require 'action_controller'+

Back to _actionpack/lib/action_controller.rb_.

After the initial call to +require 'abstract_controller'+, this calls +require 'action_dispatch'+ which was required earlier by _railties/lib/rails.rb_. The purpose of this file is explained in the ActionDispatch Railtie section.

This file defines the +ActionController+ module and its autoloaded classes.

Here we have a new method called +autoload_under+. This was covered in the Active Record Railtie but it is covered here also just in case you missed or skimmed over it. The +autoload_under+ method is  defined in +ActiveSupport::Autoload+ and it sets the autoload path to temporarily be the specified path, in this case by specifying _metal_ it will load the specified +autoload+'d classes from _lib/action_controller/metal_ inside the block.

Another new method we have here is called +autoload_at+:

<ruby>
  autoload_at "action_controller/metal/exceptions" do
    autoload :ActionControllerError
    autoload :RenderError
    autoload :RoutingError
    autoload :MethodNotAllowed
    autoload :NotImplemented
    autoload :UnknownController
    autoload :MissingFile
    autoload :RenderError
    autoload :SessionOverflowError
    autoload :UnknownHttpMethod
  end
</ruby>

This defines the path of which to find these classes defined at and is most useful for if you have multiple classes defined in a single file, as is the case for this block; all of those classes are defined inside _action_controller/metal/exceptions.rb_ and when Active Support goes looking for them it will look in that file.

At the end of this file there are a couple more requires:

<ruby>
  # All of these simply register additional autoloads
  require 'action_view'
  require 'action_controller/vendor/html-scanner'

  # Common Active Support usage in ActionController
  require 'active_support/concern'
  require 'active_support/core_ext/class/attribute_accessors'
  require 'active_support/core_ext/load_error'
  require 'active_support/core_ext/module/attr_internal'
  require 'active_support/core_ext/module/delegation'
  require 'active_support/core_ext/name_error'
  require 'active_support/inflector'
</ruby>

h5. +require 'action_view'+

This is best covered in the Action View Railtie section, so skip there by TODO: Link / page?

h5. +require 'action_controller/vendor/html-scanner'+

TODO: What is the purpose of this? Find out.

h5. +require 'active_support/concern'+

TODO: I can kind of understand the purpose of this.. need to see where @_dependencies is used however.

h5. +require 'active_support/core_ext/class/attribute_accessors'+

This file defines, as the path implies, attribute accessors for class. These are +cattr_reader+, +cattr_writer+, +cattr_accessor+.

h5. +require 'active_support/core_ext/load_error'+

The Active Support Core Extensions (TODO: LINK!) guide has a great coverage of what this file precisely provides.

h5. +require 'active_support/core_ext/module/attr_internal'+

This file is explained in the "Core Extension" guide.

This file was required through the earlier _abstract_controller.rb_ require.

h5. +require 'active_support/core_ext/module/delegation'+

This file is explained in the "Common Includes" section.

This file was required earlier by Arel and so is not required again.

h5. +require 'active_support/core_ext/name_error'+

This file includes extensions to the +NameError+ class, providing the +missing_name+ and +missing_name?+ methods. For more information see the Active Support Core Extensions guide.

h5. +require 'active_support/inflector'+

This file is explained in the "Common Includes" section.

This file was earlier required by Arel and so is not required again.

h5. Action Controller Railtie

So now we come back to the Action Controller Railtie with a couple more requires to go before +ActionController::Railtie+ is defined:

<ruby>
  require "action_view/railtie"
  require "active_support/core_ext/class/subclasses"
  require "active_support/deprecation/proxy_wrappers"
  require "active_support/deprecation"
</ruby>

As explained previously the +action_view/railtie+ file will be explained in the Action View Railtie section. TODO: link to it.

h5. +require 'active_support/core_ext/class/subclasses'+

For an explanation of this file _activesupport/lib/active_support/core_ext/class/subclasses_, see the Active Support Core Extension guide.

h5. +require 'active_support/deprecation/proxy_wrappers'+

This file, _activesupport/lib/active_support/deprecation/proxy_wrappers.rb_, defines a couple of deprecation classes, which are +DeprecationProxy+, +DeprecationObjectProxy+, +DeprecationInstanceVariableProxy+, +DeprecationConstantProxy+ which are all namespaced into +ActiveSupport::Deprecation+. These last three are all subclasses of +DeprecationProxy+.

Why do we mention them here? Beside the obvious-by-now fact that we're covering just about everything about the initialization process in this guide, if you're deprecating something in your library and you use Active Support, you too can use the +DeprecationProxy+ class (and it's subclasses) too.


h6. +DeprecationProxy+

This class is used only in _railties/lib/rails/deprecation.rb_, loaded further on in the initialization process. It's used in this way:

<ruby>
  RAILS_ROOT = (Class.new(ActiveSupport::Deprecation::DeprecationProxy) do
    cattr_accessor :warned
    self.warned = false

    def target
      Rails.root
    end

    def replace(*args)
      warn(caller, :replace, *args)
    end

    def warn(callstack, called, args)
      unless warned
        ActiveSupport::Deprecation.warn("RAILS_ROOT is deprecated! Use Rails.root instead", callstack)
        self.warned = true
      end
    end
  end).new
</ruby>

There is similar definitions for the other constants of +RAILS_ENV+ and +RAILS_DEFAULT_LOGGER+. All three of these constants are in the midst of being deprecated (most likely in Rails 3.1) so Rails will tell you if you reference them that they're deprecated using the +DeprecationProxy+ class. Whenever you call +RAILS_ROOT+ this will raise a warning, telling you: "RAILS_ROOT is deprecated! Use Rails.root instead".... TODO: investigate if simply calling it does raise this warning. This same rule applies to +RAILS_ENV+ and +RAILS_DEFAULT_LOGGER+, their new alternatives are +Rails.env+ and +Rails.logger+ respectively.

h6. +DeprecatedObjectProxy+

This is used in one place _actionpack/lib/action_controller/railtie.rb_, which you may remember is how we got to the +DeprecationProxy+ section:

<ruby>
   ActiveSupport::Deprecation::DeprecatedObjectProxy.new(app.routes, message)
</ruby>

This makes more sense in the wider scope of the initializer:

<ruby>
  initializer "action_controller.url_helpers" do |app|
    ActionController.base_hook do
      extend ::ActionController::Railtie::UrlHelpers.with(app.routes)
    end

    message = "ActionController::Routing::Routes is deprecated. " \
              "Instead, use Rails.application.routes"

    proxy = ActiveSupport::Deprecation::DeprecatedObjectProxy.new(app.routes, message)
    ActionController::Routing::Routes = proxy
  end
</ruby>

+ActionController::Routing::Routes+ was the previous constant used in defining routes in Rails 2 applications, now it's simply a method on +Rails.application+ rather than it's own individual class: +Rails.application.routes+. Both of these still call the +draw+ method on the returned object to end up defining the routes.


h6. +DeprecatedInstanceVariableProxy+

This isn't actually used anywhere in Rails anymore. It was used previously for when +@request+ and +@params+ were deprecated in Rails 2. It has been kept around as it could be useful for the same purposes in libraries that use Active Support.

h6. +DeprecatedConstantProxy+

This method is used in a couple of places, _activesupport/lib/active_support/json/encoding.rb_ and _railties/lib/rails/rack.rb_.

In _encoding.rb_ it's used to define a constant that's now been deprecated:

<ruby>
  CircularReferenceError = Deprecation::DeprecatedConstantProxy.new('ActiveSupport::JSON::CircularReferenceError', Encoding::CircularReferenceError)
</ruby>


Now when you reference +ActiveSupport::JSON::CircularReferenceError+ you'll receive a warning:

<plain>
  ActiveSupport::JSON::CircularReferenceError is deprecated! Use Encoding::CircularReferenceError instead.
</plain>

h5. +require "active_support/deprecation"+

This re-opens the +ActiveSupport::Deprecation+ module which was already defined by our deprecation proxies. Before this happens however we have 4 requires:

<ruby>
  require 'active_support/deprecation/behaviors'
  require 'active_support/deprecation/reporting'
  require 'active_support/deprecation/method_wrappers'
  require 'active_support/deprecation/proxy_wrappers'
</ruby>

The remainder of this file goes about setting up the +silenced+ and +debug+ accessors:

<ruby>
  module ActiveSupport
    module Deprecation #:nodoc:
      class << self
        # The version the deprecated behavior will be removed, by default.
        attr_accessor :deprecation_horizon
      end
      self.deprecation_horizon = '3.0'

      # By default, warnings are not silenced and debugging is off.
      self.silenced = false
      self.debug = false
    end
  end
</ruby>

h5. +require "active_support/deprecation/behaviors"+

This sets up some default behavior for the warnings raised by +ActiveSupport::Deprecation+, defining different ones for _development_ and _test_ and nothing for production, as we never want deprecation warnings in production:

<ruby>
  # Default warning behaviors per Rails.env. Ignored in production.
  DEFAULT_BEHAVIORS = {
    'test' => Proc.new { |message, callstack|
       $stderr.puts(message)
       $stderr.puts callstack.join("\n  ") if debug
     },
    'development' => Proc.new { |message, callstack|
       logger =
         if defined?(Rails) && Rails.logger
           Rails.logger
         else
           require 'logger'
           Logger.new($stderr)
         end
       logger.warn message
       logger.debug callstack.join("\n  ") if debug
     }
  }
</ruby>

In the _test_ environment, we will see the deprecation errors displayed in +$stderr+ and in _development_ mode, these are sent to +Rails.logger+ if it exists, otherwise it is output to +$stderr+ in a very similar fashion to the _test_ environment. These are both defined as procs, so Active Support can pass arguments to the +call+ method we call on it when Active Support +warn+.

h5. +require 'active_support/deprecation/reporting'+

This file defines further extensions to the +ActiveSupport::Deprecation+ module, including the +warn+ method which is used from Active Support's +DeprecationProxy+ class and an +attr_accessor+ on the class called +silenced+. This checks that we have a behavior defined, which we do in the _test_ and _development_ environments, and that we're not +silenced+ before warning about deprecations by +call+'ing the +Proc+ time.

This file also defines a +silence+ method on the module also which you can pass a block to temporarily silence errors:

<ruby>
  ActiveSupport::Deprecation.silence do
    puts "YOU CAN FIND ME HERE: #{RAILS_ROOT}"
  end
</ruby>

TODO: may have to correct this example.

h5. +require 'active_support/deprecation/method_wrappers'+

This file defines a class method on +ActiveSupport::Deprecation+ called +deprecate_methods+. This method is used in _activesupport/lib/active_support/core_ext/module/deprecation.rb_ to allow you to declare deprecated methods on modules:

<ruby>
  class Module
    # Declare that a method has been deprecated.
    #   deprecate :foo
    #   deprecate :bar => 'message'
    #   deprecate :foo, :bar, :baz => 'warning!', :qux => 'gone!'
    def deprecate(*method_names)
      ActiveSupport::Deprecation.deprecate_methods(self, *method_names)
    end
  end
</ruby>

h5. +require 'action_controller/railtie'+

Inside +ActionController::Railtie+ there are another two requires:

<ruby>
  require "action_controller/railties/log_subscriber"
  require "action_controller/railties/url_helpers"
</ruby>


h5. +require 'action_controller/railties/log_subscriber'+

+ActionController::Railties::LogSubscriber+ inherits from +Rails::LogSubscriber+ and defines methods for logging such things as action processing and file sending.

h5. +require 'action_controller/railties/url_helpers'+

This file defines a +with+ method on +ActionController::Railtie::UrlHelpers+ which is later used in the +action_controller.url_helpers+ initializer. For more information see the +action_controller.url_helpers+ initializer section.

h5. Action Controller Railtie

After these requires it deprecates a couple of ex-Action Controller methods and points whomever references them to their ActionDispatch equivalents. These methods are +session+, +session=+, +session_store+ and +session_store=+.

After the deprecations, Rails defines the +log_subscriber+ to be a new instance of +ActionController::Railties::LogSubscriber+ and then go about defining the following initializers, keeping in mind that these are added to the list of initializers defined before hand:

* action_controller.logger
* action_controller.set_configs
* action_controller.initialize_framework_caches
* action_controller.set_helpers_path
* action_controller.url_helpers

h4. Action View Railtie

The Action View Railtie provides the backend code for your views and it puts the C into MVC. This implements the +ActionView::Base+ of which all views and partials are objects of.

h5. +require 'action_view/railtie'+

The Railtie is defined in a file called _actionpack/lib/action_view/railtie.rb_ and initially makes a call to +require 'action_view'+.

h5. +require 'action_view'+

Here again we have the addition of the path to Active Support to the load path attempted, but because it's already in the load path it will not be added. Similarly, we have two requires:

<ruby>
  require 'active_support/ruby/shim'
  require 'active_support/core_ext/class/attribute_accessors'
</ruby>

And these have already been required. If you wish to know what these files do go to the explanation of each in the "Common Includes" section. TODO: link to them!

This file goes on to +require 'action_pack'+ which consists of all this code (comments stripped):

<ruby>
  require 'action_pack/version'
</ruby>

the _version_ file contains this code (comments stripped):

<ruby>
  module ActionPack #:nodoc:
    module VERSION #:nodoc:
      MAJOR = 3
      MINOR = 1
      TINY  = 0
      BUILD = "beta"

      STRING = [MAJOR, MINOR, TINY, BUILD].join('.')
    end
  end
</ruby>

TODO: Why?!

This file goes on to define the +ActionView+ module and its +autoload+'d modules and then goes on to make two more requires:

<ruby>
  require 'active_support/core_ext/string/output_safety'
  require 'action_view/base'
</ruby>

h5. +require 'active_support/core_ext/string/output_safety'+

The _actionpack/lib/active_support/core_ext/string/output_saftey.rb_ file is responsible for the code used in escaping HTML and JSON, namely the +html_escape+ and +json_escape+ methods. It does this by overriding these methods in +Erb::Util+ which is later included into +ActionView::Base+. This also defines +ActiveSupport::SafeBuffer+ which descends from +String+ and is used for concatenating safe output from your views to ERB templates.

h5. +require 'action_view/base'+

This file initially makes requires to the following files:

<ruby>
  require 'active_support/core_ext/module/attr_internal'
  require 'active_support/core_ext/module/delegation'
  require 'active_support/core_ext/class/attribute'
</ruby>

These are explained in their relevant areas inside the "Common Includes" section.

The remainder of this file sets up the +ActionView+ module and the +ActionView::Base+ class which is the class of all view templates. Inside of +ActionView::Base+ it makes an include to several helper modules:

<ruby>
  include Helpers, Rendering, Partials, Layouts, ::ERB::Util, Context
</ruby>

h5. +ActionView::Helpers+

This module, from _actionpack/lib/action_view/helpers.rb_, initially sets up the +autoload+'s for the various +ActionView::Helpers+ modules (TODO: mysteriously not using +autoload_under+). This also sets up a +ClassMethods+ module which is included automatically into wherever +ActionView::Helpers+ is included by defining a +self.included+ method:

<ruby>
  def self.included(base)
    base.extend(ClassMethods)
  end

  module ClassMethods
    include SanitizeHelper::ClassMethods
  end
</ruby>

Inside of +SanitizeHelper::ClassMethods+ it defines, of course, methods for assisting with sanitizing in Rails such as +link_sanitizer+ which is used by the +strip_links+ method.

Afterwards this includes the +ActiveSupport::Benchmarkable+ which is used for benchmarking how long a specific thing takes in a view. The method is simply +benchmark+ and can be used like this:

<ruby>
  benchmark("potentially long running thing") do
    Post.count
  end
</ruby>

The "documentation":http://api.rubyonrails.org/classes/ActiveSupport/Benchmarkable.html is great about explaining what precisely this does.

This module is also included into Active Record and +AbstractController+, meaning you can also use the +benchmark+ method in these methods.

After including +ActiveSupport::Benchmarkable+, the helpers which we have declared to be +autoload+'d are included. I will not go through and cover what each of these helpers do, as their names should be fairly explicit about it, and it's not really within the scope of this guide.

h5. +ActionView::Rendering+

This module, from _actionpack/lib/action_view/render/rendering.rb_ defines a method you may be a little too familiar with: +render+. This is the +render+ use for rendering all kinds of things, such as partials, templates and text.

h5. +ActionView::Partials+

This module, from _actionpack/lib/action_view/render/partials.rb_, defines +ActionView::Partials::PartialRenderer+ which you can probably guess is used for rendering partials.

h5. +ActionView::Layouts+

This module, from _actionpack/lib/action_view/render/layouts.rb_, defines +ActionView::Layouts+ which defines methods such as +find_layout+ for locating layouts.

h5. +ERB::Util+

The +ERB::Util+ module from Ruby core, as the document describes it: "A utility module for conversion routines, often handy in HTML generation". It offers two methods +html_escape+ and +url_encode+, with a third called +json_escape+ being added in by the requirement of _actionpack/lib/active_support/core_ext/string/output_saftey.rb_ earlier. As explained earlier, +html_escape+ is overridden to return a string marked as safe.

h5. +ActionView::Context+

TODO: Not entirely sure what this is all about. Something about the context of view rendering... can't work it out.

h5. Action View Railtie

Now that _actionpack/lib/action_view.rb_ has been required, the next step is to +require 'rails'+, but this will be skipped as the file was required by _railties/lib/rails/all.rb_ way back in the beginnings of the initialization process.

Next, the Railtie itself is defined:


<ruby>
  module ActionView
    class Railtie < Rails::Railtie
      railtie_name :action_view

      require "action_view/railties/log_subscriber"
      log_subscriber ActionView::Railties::LogSubscriber.new

      initializer "action_view.cache_asset_id" do |app|
        unless app.config.cache_classes
          ActiveSupport.on_load(:action_view) do
            ActionView::Helpers::AssetTagHelper::AssetPaths.cache_asset_ids = false
          end
        end
      end
    end
  end
</ruby>

The +ActionView::LogSubscriber+ sets up a method called +render_template+ which is called when a template is rendered. TODO: Templates only or partials and layouts also? I would imagine these fall under the templates category, but there needs to research to ensure this is correct.

The sole initializer defined here, _action_view.cache_asset_ids_ is responsible for caching the timestamps on the ends of your assets. If you've ever seen a link generated by +image_tag+ or +stylesheet_link_tag+ you would know that I mean that this timestamp is the number after the _?_ in this example: _/javascripts/prototype.js?1265442620_. This initializer will do nothing if +cache_classes+ is set to false in any of your application's configuration. TODO: Elaborate.

h4. Action Mailer Railtie

The Action Mailer Railtie is responsible for including all the emailing functionality into Rails by way of the Action Mailer gem itself. Action Mailer is:

Action Mailer is a framework for designing email-service layers. These layers
are used to consolidate code for sending out forgotten passwords, welcome
wishes on signup, invoices for billing, and any other use case that requires
a written notification to either a person or another system.

Action Mailer is in essence a wrapper around Action Controller and the
Mail gem.  It provides a way to make emails using templates in the same
way that Action Controller renders views using templates.

TODO: Quotify.

h5. +require 'action_mailer/railtie'+

This file first makes two requires:

<ruby>
  require "action_mailer"
  require "rails"
</ruby>

The requires in +action_mailer+ are already loaded or are core extensions:

<ruby>
  require 'abstract_controller'
  require 'action_view'

  # Common Active Support usage in Action Mailer
  require 'active_support/core_ext/class'
  require 'active_support/core_ext/object/blank'
  require 'active_support/core_ext/array/uniq_by'
  require 'active_support/core_ext/module/attr_internal'
  require 'active_support/core_ext/module/delegation'
  require 'active_support/core_ext/string/inflections'
  require 'active_support/lazy_load_hooks'
</ruby>

_abstract_controller_ is covered in the "Action Controller Railtie" section. TODO: Cover AbstractController there and link to it.
_action_view_ was required by the Action View Railtie and will not be required again.

For the core extensions you may reference the "Core Extensions" guide. TODO: Link to guide.

_active_support/lazy_load_hooks_ was covered earlier in the guide and since it has already been required at this point in the initialization process, it will not be required again.

The +require "rails"+ is referencing the _railties/lib/rails.rb_ file which was included back in TODO: link to section.

_actionmailer/lib/action_mailer.rb_ then goes on to define the +ActionMailer+ module:

<ruby>
  module ActionMailer
    extend ::ActiveSupport::Autoload

    autoload :AdvAttrAccessor
    autoload :Collector
    autoload :Base
    autoload :DeliveryMethods
    autoload :DeprecatedApi
    autoload :MailHelper
    autoload :OldApi
    autoload :Quoting
    autoload :TestCase
    autoload :TestHelper
  end
</ruby>

And a +Text+ module too:

<ruby>
  module Text
    extend ActiveSupport::Autoload

    autoload :Format, 'text/format'
  end
</ruby>

which is used by the +ActionMailer::MailerHelper+ method +block_format+:

<ruby>
  def block_format(text)
    formatted = text.split(/\n\r\n/).collect { |paragraph|
      Text::Format.new(
        :columns => 72, :first_indent => 2, :body_indent => 2, :text => paragraph
      ).format
    }.join("\n")

    # Make list points stand on their own line
    formatted.gsub!(/[ ]*([*]+) ([^*]*)/) { |s| "  #{$1} #{$2.strip}\n" }
    formatted.gsub!(/[ ]*([#]+) ([^#]*)/) { |s| "  #{$1} #{$2.strip}\n" }

    formatted
  end
</ruby>

h5. Action Mailer Railtie

The Railtie defines the +log_subscriber+ as +ActionMailer::Railties::LogSubscriber.new+, with this class having two logging methods: one for delivery called +deliver+ and one for receipt called +receive+.

The initializers defined in this Railtie are:

* action_mailer.url_for
* action_mailer.logger
* action_mailer.set_configs

These are covered later on the Initialization section. TODO: first write then link to Initialization section.

h4. Active Resource Railtie

The Active Resource Railtie is responsible for creating an interface into remote sites that offer a REST API. The Active Resource Railtie depends on Active Support and Active Model.

h5. +require 'active_resource/railtie'+

This file defines the Active Resource Railtie:

<ruby>
  require "active_resource"
  require "rails"

  module ActiveResource
    class Railtie < Rails::Railtie
      railtie_name :active_resource

      require "active_resource/railties/log_subscriber"
      log_subscriber ActiveResource::Railties::LogSubscriber.new

      initializer "active_resource.set_configs" do |app|
        app.config.active_resource.each do |k,v|
          ActiveResource::Base.send "#{k}=", v
        end
      end
    end
  end
</ruby>

The +require 'rails'+ has already been done back in TODO: link to section.

h5. +require 'active_resource'+

This file, _activeresource/lib/active_resource.rb_, defines the +ActiveResource+ module, first off this will add the path to Active Support and Active Model to the load path if it's not already there, then require both +active_support+ (_activesupport/lib/active_support.rb_) and +active_model+ (_activemodel/lib/active_model.rb_)

<ruby>
  activesupport_path = File.expand_path('../../../activesupport/lib', __FILE__)
  $:.unshift(activesupport_path) if File.directory?(activesupport_path) && !$:.include?(activesupport_path)

  activemodel_path = File.expand_path('../../../activemodel/lib', __FILE__)
  $:.unshift(activemodel_path) if File.directory?(activemodel_path) && !$:.include?(activemodel_path)

  require 'active_support'
  require 'active_model'

  module ActiveResource
    extend ActiveSupport::Autoload

    autoload :Base
    autoload :Connection
    autoload :CustomMethods
    autoload :Formats
    autoload :HttpMock
    autoload :Observing
    autoload :Schema
    autoload :Validations
  end
</ruby>

h5. Active Resource Railtie

The Railtie itself is fairly short as Active Resource is the smallest component of Rails.

<ruby>
  module ActiveResource
    class Railtie < Rails::Railtie
      railtie_name :active_resource

      require "active_resource/railties/log_subscriber"
      log_subscriber ActiveResource::Railties::LogSubscriber.new

      initializer "active_resource.set_configs" do |app|
        app.config.active_resource.each do |k,v|
          ActiveResource::Base.send "#{k}=", v
        end
      end
    end
  end
</ruby>

The Railtie defines the +log_subscriber+ as +ActiveResource::Railties::LogSubscriber.new+ which has one method defined: +request+. +request+ is used whenever a request is made to an external service.

There is only one initializer defined here: +set_configs+. This is covered later in the Initialization section.


h4. ActionDispatch Railtie

ActionDispatch handles all dispatch work for Rails. It interfaces with Action Controller to determine what action to undertake when a request comes in. TODO: I would quote the README but it is strangely absent. Flyin' blind here!

The ActionDispatch Railtie was previously required when we called +require 'rails'+, but we will cover the Railtie here too.

ActionDispatch depends on Active Support.

h5. +require 'action_dispatch/railtie'+

This file defines the ActionDispatch Railtie:

<ruby>
  require "action_dispatch"
  require "rails"

  module ActionDispatch
    class Railtie < Rails::Railtie
      railtie_name :action_dispatch

      config.action_dispatch.x_sendfile_header = "X-Sendfile"
      config.action_dispatch.ip_spoofing_check = true

      # Prepare dispatcher callbacks and run 'prepare' callbacks
      initializer "action_dispatch.prepare_dispatcher" do |app|
        # TODO: This used to say unless defined?(Dispatcher). Find out why and fix.
        require 'rails/dispatcher'
        ActionDispatch::Callbacks.to_prepare { app.routes_reloader.reload_if_changed }
      end
    end
  end
</ruby>

The +require 'rails'+ has already been done back in TODO: link to section.




h5. +require 'action_dispatch'+

This file was already loaded earlier in the initialization process. TODO: link to it.

h5. ActionDispatch Railtie

The ActionDispatch Railtie is almost as short as the Active Resource Railtie:

<ruby>
  require "action_dispatch"
  require "rails"

  module ActionDispatch
    class Railtie < Rails::Railtie
      railtie_name :action_dispatch

      config.action_dispatch.x_sendfile_header = "X-Sendfile"
      config.action_dispatch.ip_spoofing_check = true

      # Prepare dispatcher callbacks and run 'prepare' callbacks
      initializer "action_dispatch.prepare_dispatcher" do |app|
        # TODO: This used to say unless defined?(Dispatcher). Find out why and fix.
        require 'rails/dispatcher'
        ActionDispatch::Callbacks.to_prepare { app.routes_reloader.reload_if_changed }
      end
    end
  end
</ruby>

The +config+ method here is from +Rails::Railtie+ and pertains to your application's configuration. In this case, it is setting up some defaults which you can later override in _config/application.rb_.

This Railtie does not define a +log_subscriber+ and only defines one initializer: +prepare_dispatcher+.

h3. Return to _config/application.rb_

Now that Rails has finished loading all the Railties by way of +require 'rails/all'+ Rails can now move on to the next line:

<ruby>
  Bundler.require :default, Rails.env
</ruby>

NOTE: It is worth mentioning here that you are not tied to using Bundler with Rails 3, but it is (of course) advised that you do. To "turn off" Bundler, comment out or remove the corresponding lines in _config/application.rb_ and _config/boot.rb_.

Bundler was +require+'d back in _config/boot.rb_, and so that is what makes it available here. This guide does not dive into the internals of Bundler; it's really it's own separate guide.

The +Bundler.require+ method adds all the gems not specified inside a +group+ in the +Gemfile+ and the ones specified in groups for the +Rails.env+ (in this case, _development_), to the load path. This is how an application is able to find them.

The rest of this file is spent defining your application's main class. This is it without the comments:

<ruby>
  module YourApp
    class Application < Rails::Application
      config.encoding = "utf-8"
      config.filter_parameters += [:password]
    end
  end
</ruby>

h3. Return to Rails

On the surface, this looks like a simple class inheritance. There's more underneath though. back in +Rails::Application+, the +inherited+ method is defined:

<ruby>
  def inherited(base)
    raise "You cannot have more than one Rails::Application" if Rails.application
    super
    Rails.application = base.instance
  end
</ruby>

We do not already have a +Rails.application+, so instead this resorts to calling +super+. +Rails::Application+ descends from +Rails::Engine+ and so will call the +inherited+ method in +Rails::Engine+ (in _railties/lib/rails/engine.rb_), but before that it's important to note that +called_from+ is defined an +attr_accessor+ on +Rails::Engine+ and that +YourApp::Application+ is not an +abstract_railtie+:

<ruby>
  def inherited(base)
    unless base.abstract_railtie?
      base.called_from = begin
        # Remove the line number from backtraces making sure we don't leave anything behind
        call_stack = caller.map { |p| p.split(':')[0..-2].join(':') }
        File.dirname(call_stack.detect { |p| p !~ %r[railties[\w\-\.]*/lib/rails|rack[\w\-\.]*/lib/rack] })
      end
    end

    super
  end
</ruby>

This +called_from+ setting looks a little overwhelming to begin with, but the short end of it is that it returns your application's root, something like: _/home/you/yourapp_. After +called_from+ has been set, +super+ is again called and this means the +Rails::Railtie#inherited+ method (in _railties/lib/rails/railtie.rb_):

<ruby>
  def inherited(base)
    unless base.abstract_railtie?
      base.send(:include, self::Configurable)
      subclasses << base
    end
  end
</ruby>

Again, +YourApp::Application+ will return false for +abstract_railtie+ and so the code inside the +unless+ will be ran. The first line:

<ruby>
  base.send(:include, self::Configurable)
</ruby>

includes the +self::Configurable+ module, with self being +Rails::Application+ in this context:

<ruby>
  module Rails
    class Application
      module Configurable
        def self.included(base)
          base.extend ClassMethods
        end

        module ClassMethods
          def inherited(base)
            raise "You cannot inherit from a Rails::Application child"
          end
        end

        def config
          @config ||= Application::Configuration.new(self.class.find_root_with_flag("config.ru", Dir.pwd))
        end
      end
    end
  end
</ruby>

The inclusion of the +Rails::Application::Configurable+ module triggers the +included+ method in here which extends +YourApp::Application+ with the +Rails::Application::Configurable::ClassMethods+.

Now that the chain of +super+ calls is done, we'll go back to the original +inherited+ method in +Rails::Application+ and the final line in this method:

<ruby>
  Rails.application = base.instance
</ruby>

+base+ in this case is +YourApp::Application+ and calling +instance+ on this will return an instance of +YourApp::Application+ through the +instance+ method defined here:

<ruby>
  def instance
    if self == Rails::Application
      Rails.application
    else
      @@instance ||= new
    end
  end
</ruby>

+self+ in this case is +YourApp::Application+, so it won't match to +Rails::Application+ so instead the +new+ method is called which calls the +initialize+ method.




h3. Firing it up!

Now that we've covered the boot process of Rails the next line best to cover would be what happens after _script/rails_ has loaded _config/boot.rb_. That's quite simply that it then +require 'rails/commands'+ which is located at _railties/lib/rails/commands.rb_. Remember how +exec+ passed the arguments to +script/rails+? This is where they're used. _rails/commands.rb_ is quite a large file in Rails 3, as it contains all the Rails commands like console, about, generate and, of course, server. Because we've called +rails server+ the first argument in +ARGV+ is of course +"server"+. So assuming this we can determine that the +ARGV.shift+ in _commands.rb_ is going to return +"server"+, therefore it'll match this +when+:

<ruby>
  when 's', 'server'
  require 'rails/commands/server'
  Dir.chdir(ROOT_PATH)
  Rails::Server.start
</ruby>

The keen-eyed observer will note that this +when+ also specifies the argument could also be simply +'s'+ thereby making the full command +rails s+. This is the same with the other commands with +generate+ becoming +g+, +console+ becoming +c+ and +dbconsole+ becoming +db+.

This code here ensures we are at the +ROOT_PATH+ of our application (this constant was defined in _script/rails_) and then calls +Rails::Server.start+. +Rails::Server+ descends from +Rack::Server+ which is defined in the rack gem. The +Rails::Server.start+ method is defined like this:

<ruby>
  def start
    ENV["RAILS_ENV"] = options[:environment]

    puts "=> Booting #{ActiveSupport::Inflector.demodulize(server)}"
    puts "=> Rails #{Rails.version} application starting in #{Rails.env} on http://#{options[:Host]}:#{options[:Port]}"
    puts "=> Call with -d to detach" unless options[:daemonize]
    trap(:INT) { exit }
    puts "=> Ctrl-C to shutdown server" unless options[:daemonize]

    super
  ensure
    puts 'Exiting' unless options[:daemonize]
  end
</ruby>

We can see here that there is usual output indicating that the server is booting up.

How the +options+ variable gets set and how Rack starts the server up is covered in the next section.

h3. Racking it up!


This +Rack::Server.start+ method is defined like this:

<ruby>
  def self.start
    new.start
  end
</ruby>

+new+ as you know calls +initialize+ in a class, and that is defined like this:

<ruby>
  def initialize(options = nil)
    @options = options
  end
</ruby>

And then +options+, which are the options referenced by the +start+ method in +Rails::Server+.

<ruby>
  def options
    @options ||= parse_options(ARGV)
  end
</ruby>

And +parse_options+:

<ruby>
  def parse_options(args)
    options = default_options

    # Don't evaluate CGI ISINDEX parameters.
    # http://hoohoo.ncsa.uiuc.edu/cgi/cl.html
    args.clear if ENV.include?("REQUEST_METHOD")

    options.merge! opt_parser.parse! args
    options
  end
</ruby>

And +default_options+:

<ruby>
  def default_options
    {
      :environment => "development",
      :pid         => nil,
      :Port        => 9292,
      :Host        => "0.0.0.0",
      :AccessLog   => [],
      :config      => "config.ru"
    }
  end
</ruby>

Finally! We've arrived at +default_options+ which leads into our next point quite nicely. After the object has been +initialize+'d, +start+ is called:

<ruby>
  def start
    if options[:debug]
      $DEBUG = true
      require 'pp'
      p options[:server]
      pp wrapped_app
      pp app
    end

    if options[:warn]
      $-w = true
    end

    if includes = options[:include]
      $LOAD_PATH.unshift *includes
    end

    if library = options[:require]
      require library
    end

    daemonize_app if options[:daemonize]
    write_pid if options[:pid]
    server.run wrapped_app, options
  end
</ruby>

We're not debugging anything, so there goes the first 7 lines, we're not warning, nor are we including, requiring, daemonising or writing out a pid file. That's everything except the final line, which calls +run+ with the +wrapped_app+ which is then defined like this:

<ruby>
  def wrapped_app
    @wrapped_app ||= build_app app
  end
</ruby>

and +build_app+'s first and only argument is +app+ which is defined like this:


<ruby>
  def app
    @app ||= begin
      if !::File.exist? options[:config]
        abort "configuration #{options[:config]} not found"
      end

      app, options = Rack::Builder.parse_file(self.options[:config], opt_parser)
      self.options.merge! options
      app
    end
  end
</ruby>

+options+ is a method we talked about a short while ago, which is just the set of default options. +options[:config]+ in this context is therefore _config.ru_ which coincidentally we have in our application! To get an application instance from this method +Rack::Builder+ joins the fray with a call to +parse_file+ on our _config.ru_:

<ruby>
  def self.parse_file(config, opts = Server::Options.new)
    options = {}
    if config =~ /\.ru$/
      cfgfile = ::File.read(config)
      if cfgfile[/^#\\(.*)/] && opts
        options = opts.parse! $1.split(/\s+/)
      end
      cfgfile.sub!(/^__END__\n.*/, '')
      app = eval "Rack::Builder.new {( " + cfgfile + "\n )}.to_app",
        TOPLEVEL_BINDING, config
    else
      require config
      app = Object.const_get(::File.basename(config, '.rb').capitalize)
    end
    return app, options
  end
</ruby>

First this reads your config file and checks it for +#\+ at the beginning. This is supported if you want to pass options into the +Rack::Server+ instance that you have and can be used like this:

<ruby>
  #\\ -E production
  # This file is used by Rack-based servers to start the application.

  require ::File.expand_path('../config/environment',  __FILE__)
  run YourApp::Application.instance

</ruby>

TODO: Is the above correct? I am simply guessing!

After that it removes all the content after any +__END__+ in your _config.ru_ (TODO: because? Is this so it doesn't get eval'd?) and then evals the content of this file which, as you've seen is quite simple. The code that's first evaluated would be the require to the _config/environment.rb_ file, which leads into the next section.

h3. _config/environment.rb_

Now that we've seen that _rails/server_ gets to _config/environment.rb_ via Rack's requiring of it and Passenger requires it straight off the line. We've covered the boot process of Rails and covered the beginnings of a Rack server starting up. We have reached a common path for both _rails/server_ and Passenger now, so let's investigate what _config/environment.rb_ does.

<ruby>
  # Load the rails application
  require File.expand_path('../application', __FILE__)

  # Initialize the rails application
  YourApp::Application.initialize!

</ruby>

As you can see, there's a require in here for _config/application.rb_, and this file looks like this:


<ruby>
  module YourApp
    class Application < Rails::Application
      # Settings in config/environments/* take precedence over those specified here.
      # Application configuration should go into files in config/initializers
      # -- all .rb files in that directory are automatically loaded.

      # Add additional load paths for your own custom dirs
      # config.load_paths += %W( #{config.root}/extras )

      # Only load the plugins named here, in the order given (default is alphabetical).
      # :all can be used as a placeholder for all plugins not explicitly named
      # config.plugins = [ :exception_notification, :ssl_requirement, :all ]

      # Activate observers that should always be running
      # config.active_record.observers = :cacher, :garbage_collector, :forum_observer

      # Set Time.zone default to the specified zone and make Active Record auto-convert to this zone.
      # Run "rake -D time" for a list of tasks for finding time zone names. Default is UTC.
      # config.time_zone = 'Central Time (US & Canada)'

      # The default locale is :en and all translations from config/locales/*.rb,yml are auto loaded.
      # config.i18n.load_path += Dir[Rails.root.join('my', 'locales', '*.{rb,yml}')]
      # config.i18n.default_locale = :de

      # Configure generators values. Many other options are available, be sure to check the documentation.
      # config.generators do |g|
      #   g.orm             :active_record
      #   g.template_engine :erb
      #   g.test_framework  :test_unit, :fixture => true
      # end
    end
  end
</ruby>

These options (and their siblings) are explained in a later section. What's important to note for this file currently is that this is where the +YourApp::Application+ class is initialized and that it's a subclass of +Rails::Application+. This is the first point where your application begins to initialize Rails and as you can see all of this is configuration stuff which your initializers and really, the rest of your application will depend on. These options and what they do will be covered later.


h3. Rails Initialization Process

Now begins the actual initialization of Rails. Previously we have covered how _rails server_ and Passenger get to this stage and the parts of Rails that they have both loaded.

h3. +Rails::Application+

The first steps for the initialization process of Rails begins when +YourApp::Application+ descends from +Rails::Application+. The +Rails::Application+ class descends from +Rails::Engine+ class which itself descends from +Rails::Railtie+ defined in _railties/lib/rails/railtie.rb_. Along this fantastical chain of superclasses, there's defined a couple of inherited class methods. These methods just so happen to be called when a class inherits from (aka: is made a subclass of) this class. This first one is for +Rails::Application+:

<ruby>
  def inherited(base)
    raise "You cannot have more than one Rails::Application" if Rails.application
    super
    Rails.application = base.instance
  end
</ruby>

This goes up the chain by using +super+ to calling +Rails::Engine.inherited+:

<ruby>
  def inherited(base)
    unless abstract_railtie?(base)
      base.called_from = begin
        call_stack = caller.map { |p| p.split(':').first }
        File.dirname(call_stack.detect { |p| p !~ %r[railties/lib/rails|rack/lib/rack] })
      end
    end

    super
  end
</ruby>

+called_from+ references where this code was called from. This is covered later on in the "Bootstrap Initializers" section.

Which then calls +Rails::Railtie.inherited+:

<ruby>
  def inherited(base)
    unless abstract_railtie?(base)
      base.send(:include, self::Configurable)
      subclasses << base
    end
  end
</ruby>

This +inherited+ first includes the +Rails::Configurable+ module on +base+, which is +YourApp::Application+. This module defines the +config+ method on +YourApp::Application+, and now it's starting to come together. You may notice that in your +config/application.rb+ file there's a +config+ method called there. This is the method from +Rails::Configurable+.

Then this adds to +Rails::Railtie.subclasses+ your application's class because... TODO: explain.

With +Rails::Railtie.inherited+ out of the way, and that being the last thing to do in +Rails::Engine.inherited+ we return to +Rails::Application.inherited+ which calls the following:

<ruby>
    Rails.application = base.instance
</ruby>

As you already know, +base+ is +YourApp::Application+ and now it's calling the +instance+ method on it. This method is defined in +Rails::Application+ like this:

<ruby>
  def instance
    if self == Rails::Application
      Rails.application
    else
      @@instance ||= new
    end
  end
</ruby>

The +new+ method here simply creates a new +Rails::Application+ and sets it to the +@@instance+ class variable. No magic.

h3. Your Application's Configuration

Now that +inherited+ has finished doing its job, next up in _config/application.rb_ is the call to the +config+ object's methods. As explained before, this +config+ object is an instance of +Rails::Railtie::Configuration+, put into place by the call of +include Rails::Configurable+ back in +Rails::Railtie.inherited+. This defined it as such:

<ruby>
  def config
    @config ||= Railtie::Configuration.new
  end
</ruby>

All the methods for +Rails::Railtie::Configuration+ are defined like this in _railties/lib/rails/railtie/configuration.rb_:

<ruby>
  require 'rails/configuration'

  module Rails
    class Railtie
      class Configuration
        include Rails::Configuration::Shared
      end
    end
  end
</ruby>

As you can probably guess here, the +Rails::Configuration+ module is defined by _rails/configuration_ (_railties/lib/rails/configuration.rb_).

h3. +Rails::Configuration::Shared+

In a standard application, the +application.rb+ looks like this with all the comments stripped out:

<ruby>
  require File.expand_path('../boot', __FILE__)

  module YourApp
    class Application < Rails::Application
      config.filter_parameters << :password
    end
  end
</ruby>

The +config+ method being the one defined on +Rails::Application::Configurable+:

<ruby>
  def config
    @config ||= Application::Configuration.new(self.class.find_root_with_flag("config.ru", Dir.pwd))
  end
</ruby>

The method +find_with_root_flag+ is defined on +Rails::Engine+ (the superclass of +Rails::Application+) and it will find the directory containing a certain flag. In this case it's the +config.ru+ file:

<ruby>
  def find_root_with_flag(flag, default=nil)
    root_path = self.called_from

    while root_path && File.directory?(root_path) && !File.exist?("#{root_path}/#{flag}")
      parent = File.dirname(root_path)
      root_path = parent != root_path && parent
    end

    root = File.exist?("#{root_path}/#{flag}") ? root_path : default
    raise "Could not find root path for #{self}" unless root

    RUBY_PLATFORM =~ /(:?mswin|mingw)/ ?
      Pathname.new(root).expand_path : Pathname.new(root).realpath
  end
</ruby>

+called_from+ goes through the +caller+ which is the stacktrace of the current thread, in the case of your application it would go a little like this:

<pre>
  /usr/local/lib/ruby/gems/1.9.1/gems/railties-3.0.0/lib/rails/application.rb:30:in `inherited'
  /home/you/yourapp/config/application.rb:4:in `<module:TestApp>'
  /home/you/yourapp/config/application.rb:3:in `<top (required)>'
  /usr/local/lib/ruby/gems/1.9.1/gems/activesupport-3.0.0/lib/active_support/dependencies.rb:167:in `require'
  /usr/local/lib/ruby/gems/1.9.1/gems/activesupport-3.0.0/lib/active_support/dependencies.rb:167:in `block in require'
  /usr/local/lib/ruby/gems/1.9.1/gems/activesupport-3.0.0/lib/active_support/dependencies.rb:537:in `new_constants_in'
  /usr/local/lib/ruby/gems/1.9.1/gems/activesupport-3.0.0/lib/active_support/dependencies.rb:167:in `require'
  /usr/local/lib/ruby/gems/1.9.1/gems/railties-3.0.0/lib/rails/commands.rb:33:in `<top (required)>'
  /usr/local/lib/ruby/gems/1.9.1/gems/activesupport-3.0.0/lib/active_support/dependencies.rb:167:in `require'
  /usr/local/lib/ruby/gems/1.9.1/gems/activesupport-3.0.0/lib/active_support/dependencies.rb:167:in `block in require'
  /usr/local/lib/ruby/gems/1.9.1/gems/activesupport-3.0.0/lib/active_support/dependencies.rb:537:in `new_constants_in'
  /usr/local/lib/ruby/gems/1.9.1/gems/activesupport-3.0.0/lib/active_support/dependencies.rb:167:in `require'
  /var/www/rboard/script/rails:10:in `<main>'
</pre>

+called_from+ is defined in the +inherited+ method for +Rails::Engine+ which looks like this:

<ruby>
  base.called_from = begin
    call_stack = caller.map { |p| p.split(':').first }
    File.dirname(call_stack.detect { |p| p !~ %r[railties/lib/rails|rack/lib/rack] })
  end
</ruby>

The +call_stack+ here is the +caller+ output shown previously, minus everything after the first +:+ on all the lines. The first path that matches this is _/usr/local/lib/ruby/gems/1.9.1/gems/railties-3.0.0/lib/rails_. Yours may vary slightly, but should always end in _railties-x.x.x/lib/rails_.

The code in +find_root_with_flag+ will go up this directory structure until it reaches the top, which in this case is +/+.

<ruby>
  while root_path && File.directory?(root_path) && !File.exist?("#{root_path}/#{flag}")
    parent = File.dirname(root_path)
    root_path = parent != root_path && parent
  end

  root = File.exist?("#{root_path}/#{flag}") ? root_path : default
  raise "Could not find root path for #{self}" unless root
</ruby>

TODO: What is all this for?

At the root of the system it looks for +config.ru+. TODO: Why? Obviously it's not going to find it, so it uses the +default+ option we've specified which is +Dir.pwd+ which will default to the root folder of your Rails application. This path is then passed to +Rails::Application::Configuration.new+. +Rails::Application::Configuration+ descends from +Rails::Engine::Configuration+ and the +initialize+ method goes like this:

<ruby>
  def initialize(*)
    super
    @allow_concurrency   = false
    @colorize_logging    = true
    @filter_parameters   = []
    @dependency_loading  = true
    @serve_static_assets = true
    @time_zone           = "UTC"
    @consider_all_requests_local = true
  end
</ruby>

The +super+ method here is the +initialize+ method in +Rails::Engine::Configuration+:

<ruby>
  def initialize(root=nil)
    @root = root
  end
</ruby>

Here, the +@root+ variable is assigned the path of your application and then the remainder of +Rails::Application::Configuration.initialize+ is ran, setting up a few instance variables for basic configuration, including one for +@filter_parameters+.

Now with the +config+ option set up, we can go onwards and call +filter_parameters+ on it. This +filter_parameters+ method is not defined on +Rails::Configuration::Shared+ and actually falls to the +method_missing+ defined there instead:

<ruby>
  def method_missing(name, *args, &blk)
     if name.to_s =~ config_key_regexp
       return $2 == '=' ? options[$1] = args.first : options[$1]
     end
     super
   end
</ruby>

We're not calling +filter_parameters=+, we're calling +filter_parameters+, therefore it'll be the second part of this ternary argument: +options[$1]+. The options method is defined like this:

<ruby>
  def options
    @@options ||= Hash.new { |h,k| h[k] = ActiveSupport::OrderedOptions.new }
  end
</ruby>

OrderedOptions exists... TODO: explain.


So from this we can determine that our +options+ hash now has a key for +filter_parameters+ which's value is an array consisting of a single symbol: +:password+. How this option manages to get into the +@filter_parameters+ variable defined on the +Rails::Application::Configuration.initialize+ method is explained later.

h3. Application Configured!

Now your application has finished being configured (at least in the sense of _config/application.rb_, there's more to come!) in _config/environment.rb_ the final line calls +YourApp::Application.initalize!+.

h3. Initialization begins

This is one of those magical uses of +method_missing+ which, for the purposes of debugging, is something that you don't expect to come across as often as you do and as a consequence you'll spend a good portion of an hour looking for method definitions that don't exist because +method_missing+ is taking care of it. There's some pretty crafty use of +method_missing+ all over Rails and it's encouraged to take note of its power.

+Rails::Application+ has a +method_missing+ definition which does this:

<ruby>
  def method_missing(*args, &block)
    instance.send(*args, &block)
  end
</ruby>

With our +instance+ being our already initialized by the +inherited+ method, this will just return the value of the +@@instance+ variable, a +Rails::Application+ object. Calling +initialize!+ on this method does this:

<ruby>
  def initialize!
    run_initializers(self)
    self
  end
</ruby>

The initializers it is talking about running here are the initializers for our application. The object passed in to +run_initializers+ is +YourApp::Application+.


h3. +run_initializers+

This method begins the running of all the defined initializers. In the section "The Boot Process" we covered the loading sequence of Rails before any initialization happens and during this time we saw that the +Rails::Railtie+ class includes the +Initializable+ module. As we've also seen +YourApp::Application+ is a descendant of this class, so it too has these methods.

+run_initializers+ looks like this:

<ruby>
  def run_initializers(*args)
    return if instance_variable_defined?(:@ran)
    initializers.each do |initializer|
      initializer.run(*args)
    end
    @ran = true
  end
</ruby>

Here the +initializers+ method is defined in _railties/lib/rails/application.rb_:

<ruby>
  def initializers
    initializers = Bootstrap.initializers_for(self)
    railties.all { |r| initializers += r.initializers }
    initializers += super
    initializers += Finisher.initializers_for(self)
    initializers
  end
</ruby>

h3. +Bootstrap+ initializers

The first line here references a +Bootstrap+ class we haven't seen before. Or have we? The keen-eyed observer would have spotted an +autoload+ for it at the top of +Rails::Application+:

<ruby>
  autoload :Bootstrap,      'rails/application/bootstrap'
</ruby>

Now that we've referenced that class, it will be required for us. You'll notice inside this class that there's an +include Initializable+, providing the afore-mentioned methods from this module. Inside this class a number of initializers are defined.

* load_environment_config
* load_all_active_support
* preload_frameworks
* initialize_logger
* initialize_cache
* initialize_subscriber
* set_clear_dependencies_hook
* initialize_dependency_mechanism

These are all defined using the +initializer+ method:

<ruby>
  def initializer(name, opts = {}, &blk)
    raise ArgumentError, "A block must be passed when defining an initializer" unless blk
    opts[:after] ||= initializers.last.name unless initializers.empty? || initializers.find { |i| i.name == opts[:before] }
    initializers << Initializer.new(name, nil, opts, &blk)
  end
</ruby>

The +initializers+ method defined here just references an +@initializers+ variable:

<ruby>
  def initializers
    @initializers ||= []
  end
</ruby>

As you can see from this method it will set +opts[:after]+ if there are previously defined initializers. So we can determine from this that the order our initializers are defined in is the same order that they run in, but only by default. It is possible to change this by specifying an +:after+ or +:before+ option as we will see later on. Each initializer is its own instance of the +Initializer+ class:

<ruby>
  class Initializer
    attr_reader :name, :block

    def initialize(name, context, options, &block)
      @name, @context, @options, @block = name, context, options, block
    end

    def before
      @options[:before]
    end

    def after
      @options[:after]
    end

    def run(*args)
      @context.instance_exec(*args, &block)
    end

    def bind(context)
      return self if @context
      Initializer.new(@name, context, @options, &block)
    end
  end
</ruby>

Now that +Rails::Application::Bootstrap+ has finished loading, we can continue on with our initialization. We saw that it called this:

<ruby>
  initializers = Bootstrap.initializers_for(self)
</ruby>

Calling +initializers_for+, defined like this:

<ruby>
  def initializers_for(binding)
    Collection.new(initializers_chain.map { |i| i.bind(binding) })
  end
</ruby>

The +binding+ argument here is +YourApp::Application+ and this will return a new +Initializer+ object for all the initializers in +initializers_chain+ for this particular context. +initializers_chain+ goes like this:

<ruby>
  def initializers_chain
    initializers = Collection.new
    ancestors.reverse_each do |klass|
      next unless klass.respond_to?(:initializers)
      initializers = initializers + klass.initializers
    end
    initializers
  end
</ruby>

The ancestors list is relatively short for +Rails::Application::Bootstrap+, consisting of itself and +Rails::Initializable+. Rails will go through these ancestors in reverse and check them all if they +respond_to?(:initializers)+. +Rails::Initializable+ does not and so it's skipped. +Rails::Application::Bootstrap+ of course does, and this is the list of initializers we covered earlier.

After +initializers_chain+ is finished, then they are +map+'d like this, with the +binding+ of course being +YourApp::Application+ as explained previously.

<ruby>
  def initializers_for(binding)
    Collection.new(initializers_chain.map { |i| i.bind(binding) })
  end
</ruby>

Wow. All that to cover just the first line in the +initializers+ method for +Rails::Application+.

h3. Railties Initializers

This section covers the loading of the initializers and we will go into depth for each initializer in the next section, as they make more sense explained in their chain.

The second line in +Rails::Application#initializers+:

<ruby>
  def initializers
    railties.all { |r| initializers += r.initializers }
  end
</ruby>

calls +railties+, which is defined like this:

<ruby>
  def railties
    @railties ||= Railties.new(config)
  end
</ruby>

This sets up a new +Rails::Application::Railties+ object like this:

<ruby>
  def initialize(config)
    @config = config
  end
</ruby>

And calls +all+ on it:

<ruby>
  def all(&block)
    @all ||= railties + engines + plugins
    @all.each(&block) if block
    @all
  end
</ruby>

This +all+ method executes code on all the +Rails::Railtie+ and +Rails::Engine+ subclasses, retreived by the +railties+ and +engines+ methods defined right after +all+:

<ruby>
  def railties
    @railties ||= ::Rails::Railtie.subclasses.map(&:new)
  end

  def engines
    @engines ||= ::Rails::Engine.subclasses.map(&:new)
  end
</ruby>

By default, the railties are:

* +ActiveSupport::Railtie+
* +I18n::Railtie+
* +ActionDispatch::Railtie+
* +ActionController::Railtie+
* +ActiveRecord::Railtie+
* +ActionView::Railtie+
* +ActionMailer::Railtie+
* +ActiveResource::Railtie+
* +Rails::TestUnitRailtie+

And these all descend from +Rails::Railtie+.

The default +engines+ are +[]+.

The +plugins+ method it calls is a little more complex:

<ruby>
  def plugins
    @plugins ||= begin
      plugin_names = (@config.plugins || [:all]).map { |p| p.to_sym }
      Plugin.all(plugin_names, @config.paths.vendor.plugins)
    end
  end
</ruby>

+@config.paths+ is defined in the +Rails::Application::Configuration+ like this:

<ruby>
  def paths
    @paths ||= begin
      paths = super
      paths.app.controllers << builtin_controller if builtin_controller
      paths.config.database    "config/database.yml"
      paths.config.environment "config/environments", :glob => "#{Rails.env}.rb"
      paths.log                "log/#{Rails.env}.log"
      paths.tmp                "tmp"
      paths.tmp.cache          "tmp/cache"
      paths.vendor             "vendor", :load_path => true
      paths.vendor.plugins     "vendor/plugins"

      if File.exists?("#{root}/test/mocks/#{Rails.env}")
        ActiveSupport::Deprecation.warn "\"RAILS_ROOT/test/mocks/#{Rails.env}\" won't be added " <<
          "automatically to load paths anymore in future releases"
        paths.mocks_path  "test/mocks", :load_path => true, :glob => Rails.env
      end

      paths
    end
  end
</ruby>

When we call +@config.paths.vendor.plugins+ it will return +"vendor/plugins"+.


If you've defined specific plugin requirements for your application in _config/application.rb_ by using this code:

<ruby>
  config.plugins = [:will_paginate, :by_star]
</ruby>

or specific plugin loading using a similar statement such as this next one:

<ruby>
  config.plugins = [:will_paginate, :by_star, :all]
</ruby>


Then this is where the +@config.plugins+ comes from. If you wish to load only certain plugins for your application, use the first example. If you wish to load certain plugins before the rest then the second example is what you would use.

If +config.plugins+ is not defined then +:all+ is specified in its place. Whatever the +plugin_names+ is specified as, is passed to +Plugin.all+ along with the path to the plugins, +@config.path.vendor.plugins+ (which defaults to _vendor/plugins_):

<ruby>
  def self.all(list, paths)
    plugins = []
    paths.each do |path|
      Dir["#{path}/*"].each do |plugin_path|
        plugin = new(plugin_path)
        next unless list.include?(plugin.name) || list.include?(:all)
        plugins << plugin
      end
    end

    plugins.sort_by do |p|
      [list.index(p.name) || list.index(:all), p.name.to_s]
    end
  end
</ruby>

As we can see here it will go through the paths and for every folder in _vendor/plugins_ and +initialize+ a new +Rails::Plugin+ object for each:

<ruby>
  def initialize(root)
    @name = File.basename(root).to_sym
    config.root = root
  end
</ruby>

This sets the plugin name to be the same name as the folder so the plugin located at _vendor/plugins/by\_star_'s name is +by_star+. After that, the +config+ object is initialized:

<ruby>
  def config
    @config ||= Engine::Configuration.new
  end
</ruby>

and the root of the plugin defined as that folder. The reasoning for defining a +root+ is so that the initializer called +load_init_rb+ has some place to look for this file:

<ruby>
  initializer :load_init_rb, :before => :load_application_initializers do |app|
    file   = Dir["#{root}/{rails/init,init}.rb"].first
    config = app.config
    eval(File.read(file), binding, file) if file && File.file?(file)
  end
</ruby>

A little more on that later, however.

If the plugin is not included in the list then it moves on to the next one. For all plugins included in the list (or if +:all+ is specified in the list) they are put into a +plugins+ local variable which is then sorted:

<ruby>
  plugins.sort_by do |p|
    [list.index(p.name) || list.index(:all), p.name.to_s]
  end
</ruby>

The sort order is the same order as which they appear in the +config.plugins+ setting, or in alphabetical order if there is no setting specified.

Now that we have our railties, engines, and plugins in a line we can finally get back to the +all+ code:

<ruby>
  def initializers
    railties.all { |r| initializers += r.initializers }
  end
</ruby>

This block will gather add the railties' initializers to it.

h3. Engine Initializers

The third line in this +initializers+ method:

<ruby>
    initializers += super
</ruby>

The +super+ method it's referring to is of course +Rails::Engine.initializers+, which isn't defined on the class but, as we have seen before, is defined on the +Rails::Railtie+ class it inherits from through the +Rails::Initializable+ module. Therefore we can determine the initializers to be added are now the ones defined in +Rails::Engine+.

h3. Finisher Initializers

The final set of initializers in this chain are those in +Rails::Finisher+. This involves running any after initialize code, building the middleware stack and adding the route for _rails/info/properties_.

h3. Running the Initializers

Now that we have all the initializers we can go back to the +run_initializers+ in +Rails::Initializable+:

<ruby>
  def run_initializers(*args)
    return if instance_variable_defined?(:@ran)
    initializers.each do |initializer|
      initializer.run(*args)
    end
    @ran = true
  end
</ruby>

Now we finally have all the +initializers+ we can go through them and call +run+:

<ruby>
  def run(*args)
    @context.instance_exec(*args, &block)
  end
</ruby>

You may remember that the +@context+ in this code is +YourApp::Application+ and calling +instance_exec+ on this class will make a new instance of it and execute the code within the +&block+ passed to it. This code within the block is the code from all the initializers.

h3. Bootstrap Initializers

These initializers are the very first initializers that will be used to get your application going.

h4. +load_environment_config+

<ruby>
  initializer :load_environment_config do
    require_environment!
  end
</ruby>

This quite simply makes a call to +require_environment!+ which is defined like this in +Rails::Application+:

<ruby>
  def require_environment!
    environment = config.paths.config.environment.to_a.first
    require environment if environment
  end
</ruby>

We've seen +config.paths+ before when loading the plugins and they're explained in more detail in the Bonus section at the end of this guide. +config.enviroment+ for +paths+ is defined like this:

<ruby>
  paths.config.environment "config/environments", :glob => "#{Rails.env}.rb"
</ruby>

+Rails.env+ was defined way back in the boot process when +railties/lib/rails.rb+ was required:

<ruby>
module Rails
  class << self

    ...

    def env
      @_env ||= ActiveSupport::StringInquirer.new(ENV["RAILS_ENV"] || ENV["RACK_ENV"] || "development")
    end

    ...

  end
end
</ruby>

With +ENV["RAILS_ENV"]+ and +ENV["RACK_ENV"]+ not set to anything for our server booting process, this will default to +"development"+.

Therefore the path to this config file line would look like this with a substitution made:

<ruby>
  paths.config.environment "config/environments", :glob => "development.rb"
</ruby>

This method returns a +Path+ object (which acts as an +Enumerable+).

Back to +require_environment+ now:

<ruby>
  def require_environment!
    environment = config.paths.config.environment.to_a.first
    require environment if environment
  end
</ruby>

And we've determined that +config.paths.config.environment+ is +Path+ object, and calling +to_a+ on that object calls +paths+ because it's +alias+'d at the bottom of the +Path+ class definition:

<ruby>
  alias to_a paths
</ruby>

<ruby>
  def paths
    raise "You need to set a path root" unless @root.path
    result = @paths.map do |p|
      path = File.expand_path(p, @root.path)
      @glob ? Dir[File.join(path, @glob)] : path
    end
    result.flatten!
    result.uniq!
    result
  end
</ruby>

This returns an array of files according to our +path+ and +@glob+ which are +config/environments+ and +development.rb+ respectively, therefore we can determine that:

<ruby>
  Dir[File.join(path, @glob)]
</ruby>

will return an +Array+ containing one element, +"config/enviroments/development.rb"+. Of course when we call +first+ on this Array we'll get the first element and because that exists, we now +require "config/environments/development.rb"+.

This file contains the following by default:

<ruby>
  YourApp::Application.configure do
    # Settings specified here will take precedence over those in config/environment.rb

    # In the development environment your application's code is reloaded on
    # every request.  This slows down response time but is perfect for development
    # since you don't have to restart the webserver when you make code changes.
    config.cache_classes = false

    # Log error messages when you accidentally call methods on nil.
    config.whiny_nils = true

    # Show full error reports and disable caching
    config.consider_all_requests_local       = true
    config.action_view.debug_rjs             = true
    config.action_controller.perform_caching = false

    # Don't care if the mailer can't send
    config.action_mailer.raise_delivery_errors = false
  end
</ruby>

This +configure+ method is an +alias+ of +class_eval+ on +Rails::Application+:

<ruby>
  alias   :configure :class_eval
</ruby>

therefore, the code inside of the +configure+ is evaluated within the context of +YourApp::Application+.

The +config+ object here is the same one that was set up when _config/application.rb_ was loaded, therefore the methods called in this object will fall to the +method_missing+ defined in +Rails::Configuration::Shared+:

<ruby>
  def method_missing(name, *args, &blk)
    if name.to_s =~ config_key_regexp
      return $2 == '=' ? options[$1] = args.first : options[$1]
    end
    super
  end
</ruby>

This time we are using methods ending in +\=+, so it will set the key in the +options+ to be the value specified. The first couple of options, +cache_classes+, +whiny_nils+, +consider_all_requests_local+ are just simple keys on the +options+. If you recall how options were setup then you may be able to work out how the remaining +action_view+, +action_controller+ and +action_mailer+ methods work.

Firstly, we'll cover how +config_key_regexp+ is defined:

<ruby>
  def config_key_regexp
    bits = config_keys.map { |n| Regexp.escape(n.to_s) }.join('|')
    /^(#{bits})(?:=)?$/
  end
</ruby>

And also +config_keys+:

<ruby>
  def config_keys
    (Railtie.railtie_names + Engine.engine_names).map { |n| n.to_s }.uniq
  end
</ruby>

+config_keys+ in here returns:

<ruby>
  [:active_support, :i18n, :action_dispatch, :action_view, :action_controller, :active_record, :action_mailer, :active_resource, :test_unit]
</ruby>

With all of those keys coming from +Railtie::railtie_names+. If you've elected to not load some of the frameworks here they won't be available as configuration keys, so you'll need to remove them too.

Now a reminder of how the +options+ key is defined:

<ruby>
  def options
    @@options ||= Hash.new { |h,k| h[k] = ActiveSupport::OrderedOptions.new }
  end
</ruby>

The values for these framework keys are +ActiveSupport::OrderedOptions+ objects, with the class defined like this:

<ruby>
  module ActiveSupport #:nodoc:
    class OrderedOptions < OrderedHash
      def []=(key, value)
        super(key.to_sym, value)
      end

      def [](key)
        super(key.to_sym)
      end

      def method_missing(name, *args)
        if name.to_s =~ /(.*)=$/
          self[$1.to_sym] = args.first
        else
          self[name]
        end
      end
    end
  end
</ruby>

We can determine when we call +config.action_view.debug_rjs+ it's falling back to the +method_missing+ defined on +ActiveSupport::OrderedOptions+, which ends up either setting or retrieving a key. In this case because we're using a setter, it will set the key for this hash. This completes the loading of _config/environments/development.rb_.

h4. +load_all_active_support+

This initializer does exactly what it says:

<ruby>
  initializer :load_all_active_support do
    require "active_support/all" unless config.active_support.bare
  end
</ruby>

If you don't want this to happen you can specify the +config.active_support.bare+ option to +true+ in either _config/application.rb_ or any of your environment files.

h4. +preload_frameworks+

Remember earlier how we had all that stuff +eager_autoload+'d for Active Support?

<ruby>
  initializer :preload_frameworks do
    require 'active_support/dependencies'
    ActiveSupport::Autoload.eager_autoload! if config.preload_frameworks
  end
</ruby>

This is where it gets loaded. The +eager_autoload!+ method is defined like this:

<ruby>
  def self.eager_autoload!
    @@autoloads.values.each { |file| require file }
  end
</ruby>

With +@@autoloads+ being


* load_all_active_support
* preload_frameworks
* initialize_logger
* initialize_cache
* initialize_subscriber
* set_clear_dependencies_hook
* initialize_dependency_mechanism

h4. Active Support Initializers

Active Support

**Active Support Initializers**

* active_support.initialize_whiny_nils
* active_support.initialize_time_zone

**I18n Initializers**

* i18n.initialize

The +I18n::Railtie+ also defines an +after_initialize+ which we will return to later when discussing the initializers in detail.

**Action Dispatch Initializers**

* action_dispatch.prepare_dispatcher

**Action Controller Initializers**

* action_controller.logger
* action_controller.set_configs
* action_controller.initialize_framework_caches
* action_controller.set_helpers_path

**Active Record Initializers**

* active_record.initialize_time_zone
* active_record.logger
* active_record.set_configs
* active_record.log_runtime
* active_record.initialize_database_middleware
* active_record.load_observers
* active_record.set_dispatch_hooks

**Action View Initializers **

* action_view.cache_asset_ids

**Action Mailer Initializers **

* action_mailer.logger
* action_mailer.set_configs
* action_mailer.url_for

**Active Resource Initializers**

* active_resource.set_configs

**Rails::Engine Initializers**

* set_load_path
* set_autoload_paths
* add_routing_paths


h4. +Rails::Engine.new+

The +new+ method doesn't exist, but in Ruby classes calling +new+ on the class instantiates a new instance of that class and calls the instance method +initialize+ on it. This method for +Rails::Application+ goes like this:

<ruby>
  def initialize
     require_environment
     Rails.application ||= self
     @route_configuration_files = []
   end
</ruby>

h4. +Rails::Application#require_environment+

This is not a crafty method like the previous ones, it just does as it says on the box:

<ruby>
  def require_environment
    require config.environment_path
  rescue LoadError
  end
</ruby>

The +config+ object here is actually another +delegate+'d method (along with +routes+), this time to +self.class+:

<ruby>
  delegate :config, :routes, :to => :'self.class'
</ruby>

So the method call is actually +self.class.config+.


h4. +Rails::Application.config+

Defined back inside the +class << self+ for +Rails::Application+, +config+ makes a new +Rails::Application::Configuration+ object and caches it in a variable called +@config+:

<ruby>
  def config
    @config ||= Configuration.new(Plugin::Configuration.default)
  end
</ruby>

h4. +Rails::Plugin::Configuration.default+

The +Rails::Plugin::Configuration+ class may be a bit difficult to find at first, but if you look for _plugin.rb_ in Rails, you'll find it in _railties/lib/rails/plugin.rb_. In this file, we see the following:

<ruby>
  module Rails
    class Plugin < Engine
      ...
    end
  end
</ruby>

So we note here that +Rails::Plugin+ descends from +Rails::Railtie+ and secondly we note that the class +Configuration+ is not actually included in the +Plugin+ class, but it **is** in the +Railtie+ class!

h4. +Rails::Railtie::Configuration+

We've now tracked down the +Plugin::Configuration.default+ method to being +Railtie::Configuration.default+, which is defined like this in _railties/lib/rails/configuration.rb_:

<ruby>
  class Railtie::Configuration
    def self.default
      @default ||= new
    end
    ...
  end
</ruby>

In this case we have effectively seen that it's doing Configuration.new(Configuration.new). I'll explain why.

h4. +Rails::Application::Configuration.new+

TODO: CLEAN THIS UP! This subclassing is only temporary and will probably not be separate in Rails 3. This is based solely off what the comment at the top of the Railtie::Configuration class says!

The first thing to note here is that this class is subclassed from +Railtie::Configuration+ and therefore the method here is actually +Railtie::Configuration.new+. As mentioned previously, calling +new+ will make a new object of this class and then call +initialize+ on it, which is defined like this:

<ruby>
  def initialize(base = nil)
    if base
      @options    = base.options.dup
      @middleware = base.middleware.dup
    else
      @options    = Hash.new { |h,k| h[k] = ActiveSupport::OrderedOptions.new }
      @middleware = self.class.default_middleware_stack
    end
  end
</ruby>

This method is not called with a +base+ argument for +Plugin::Configuration.default+ but it is for the +Configuration.new+ wrapped around it. We'll go for the internal one first, since that's the order Rails loads them in.

h4. +default_middleware_stack+

This method is defined like this:

<ruby>
  def self.default_middleware_stack
    ActionDispatch::MiddlewareStack.new.tap do |middleware|
      middleware.use('ActionDispatch::Static', lambda { Rails.public_path }, :if => lambda { Rails.application.config.serve_static_assets })
      middleware.use('::Rack::Lock', :if => lambda { !ActionController::Base.allow_concurrency })
      middleware.use('::Rack::Runtime')
      middleware.use('ActionDispatch::ShowExceptions', lambda { ActionController::Base.consider_all_requests_local })
      middleware.use('ActionDispatch::Notifications')
      middleware.use('ActionDispatch::Callbacks', lambda { !Rails.application.config.cache_classes })
      middleware.use('ActionDispatch::Cookies')
      middleware.use(lambda { ActionController::Base.session_store }, lambda { ActionController::Base.session_options })
      middleware.use('ActionDispatch::Flash', :if => lambda { ActionController::Base.session_store })
      middleware.use('ActionDispatch::ParamsParser')
      middleware.use('::Rack::MethodOverride')
      middleware.use('::ActionDispatch::Head')
    end
  end
</ruby>

To really understand this method we need to dig a little deeper, down into where +ActionDispatch::MiddlewareStack.new+ is defined and what in particular it does for us.

h4. +ActionDispatch::MiddlewareStack.new+

+ActionDispatch+ is our first foray outside of the +railties+ gem, as this is actually defined in the +actionpack+ part of Rails. The class definition is as important as the method:

<ruby>
  module ActionDispatch
    class MiddlewareStack < Array

      ...

      def initialize(*args, &block)
        super(*args)
        block.call(self) if block_given?
      end
    end
  end
</ruby>

When it's calling +super+ here it's actually calling +initialize+ on the Array class and from this we can determine that an +ActionDispatch::MiddlewareStack+ object is just an +Array+ object with special powers. One of those special powers is the ability to take a block, and +call+ it with +self+, meaning the block's parameter is the object itself!

h4. +ActionDispatch::MiddlewareStack.use+

Previously we saw a chunk of code that I'll re-show you stripped down:

<ruby>
  def self.default_middleware_stack
    ActionDispatch::MiddlewareStack.new.tap do |middleware|
      middleware.use('ActionDispatch::Static', lambda { Rails.public_path }, :if => lambda { Rails.application.config.serve_static_assets })
      ...
    end
  end
</ruby>

As explained in the previous section, we know that the +new+ on +ActionDispatch::MiddlewareStack+ takes a block and that block has one parameter which is the object itself. On this object we call the +use+ method to include middleware into our application. The use method simply does this:

<ruby>
  def use(*args, &block)
    middleware = Middleware.new(*args, &block)
    push(middleware)
  end
</ruby>

We'll come back to this method later on.

h4. +ActionController::Middleware.new+

This +initialize+ method also is in a class who's ancestry is important so once again I'll show the ancestry and we'll go up that particular chain:

<ruby>
  module ActionController
    class Middleware < Metal

    ...

      def initialize(app)
        super()
        @_app = app
      end
    end
  end
</ruby>

Here our method calls +super+ but with a difference: it's passing in no arguments intentionally by putting the two brackets at the end. The method called here is therefore +ActionController::Metal.initialize+.

h4. +ActionController::Metal.initialize+

This is another subclassed class, this time from +ActionController::AbstractController+ and I'm sure you can guess what that means:

<ruby>
  class Metal < AbstractController::Base

    ...

    def initialize(*)
      @_headers = {}
      super
    end
  end
</ruby>

The single +*+ in the argument listing means we can accept any number of arguments, we just don't care what they are.

h4. +AbstractController::Base.initialize+

This may be anti-climatic, but the initialize method here just returns an +AbstractController::Base+ object:

<ruby>
  # Initialize controller with nil formats.
  def initialize #:nodoc:
    @_formats = nil
  end
</ruby>

h4. +ActionDispatch::MiddlewareStack.use+

Now we're back to this method, from our foray into the depths of how +Middleware.new+ works, we've showed that it is an instance of +AbstractController::Base+. Therefore it does

TODO: ELABORATE ON THIS SECTION, including explaining what all the pieces of middleware do. Then explain how the default_middleware_stack does what it does, whatever that is.

h4. Back to +Rails::Application::Configuration.new+

Now that the first call to this method is complete (+Plugin::Configuration.default+), we can move onto the second call. Here's a refresher of what this method does:

<ruby>
  def initialize(base = nil)
    if base
      @options    = base.options.dup
      @middleware = base.middleware.dup
    else
      @options    = Hash.new { |h,k| h[k] = ActiveSupport::OrderedOptions.new }
      @middleware = self.class.default_middleware_stack
    end
  end
</ruby>

You'll note now that this method is being called now is +Configuration.new(Plugin::Configuration.default)+ and with the argument, it's going to perform differently than before, this time duplicating the +options+ and +middleware+ of the object it was passed.

TODO: Find out what purpose the +@options+ and +@middleware+ variables serve.

Finally, a +Rails::Application::Configuration+ object will be returned. On this class there are a couple of +attr_accessor+s and +attr_writer+s defined:

<ruby>
  attr_accessor :after_initialize_blocks, :cache_classes, :colorize_logging,
                :consider_all_requests_local, :dependency_loading,
                :load_once_paths, :logger, :plugins,
                :preload_frameworks, :reload_plugins, :serve_static_assets,
                :time_zone, :whiny_nils

  attr_writer :cache_store, :controller_paths,
              :database_configuration_file, :eager_load_paths,
              :i18n, :load_paths, :log_level, :log_path, :paths,
              :routes_configuration_file, :view_path
</ruby>

Along with these are a lot of helper methods, and one of them is +environment_path+:

<ruby>
  def environment_path
    "#{root}/config/environments/#{Rails.env}.rb"
  end
</ruby>

h4. Back to +Rails::Application#require_environment+

Now that we have a +Rails::Application::Configuration+ object for the +config+ method, we call the +environment_path+ which, as we've seen above, just requires the current environment file which in this case is _config/environments/development.rb_. If this file cannot be found, the +LoadError+ +require+ throws will be +rescue+'d and Rails will continue on its merry way.

h4. _config/environments/development.rb_

In a standard Rails application we have this in our _config/environments/development.rb_ file:

<ruby>
  YourApp::Application.configure do
    # Settings specified here will take precedence over those in config/environment.rb

    # In the development environment your application's code is reloaded on
    # every request.  This slows down response time but is perfect for development
    # since you don't have to restart the webserver when you make code changes.
    config.cache_classes = false

    # Log error messages when you accidentally call methods on nil.
    config.whiny_nils = true

    # Show full error reports and disable caching
    config.action_controller.consider_all_requests_local = true
    config.action_view.debug_rjs                         = true
    config.action_controller.perform_caching             = false

    # Don't care if the mailer can't send
    config.action_mailer.raise_delivery_errors = false
  end
</ruby>

It's a little bit sneaky here, but +configure+ is +alias+'d to +class_eval+ on subclasses of +Rails::Application+ which of course includes +YourApp::Application+. This means that the code inside the +configure do+ block will be evaled within the context of +YourApp::Application+. The +config+ method here is the one mentioned before: the +Rails::Application::Configuration+ object. The methods on it should look familiar too: they're the ones that had +attr_accessor+ and +attr_writer+ definitions.

The ones down the bottom, +config.action_controller+, +config.action_view+ and +config.action_mailer+ aren't defined by +attr_accessor+ or +attr_writer+, rather they're undefined methods and therefore will trigger the +method_missing+ on the +Rails::Application::Configuration+ option.

h5. config.cache_classes=

The first method call in this file, this tells Rails to not cache the classes for every request. This means for every single request Rails will reload the classes of your application. If you have a lot of classes, this will slow down the request cycle of your application. This is set to +false+ in the _development_ environment, and +true+ in the _test_ & _production_ environments.

h5. config.whiny_nils=

If this is set to +true+, like it is here in the _development_ environment, _activesupport/whiny_nil_ will be +require+'d. Have you ever seen this error:

<ruby>
  Called id for nil, which would mistakenly be 4 -- if you really wanted the id of nil, use object_id
</ruby>

Or perhaps this one?

<ruby>
  You have a nil object when you didn't expect it!
  You might have expected an instance of Array.
  The error occurred while evaluating nil.flatten!
</ruby>

If you have, then this is _activesupport/whiny_nil_ at work.


h5. The frameworks

As mentioned before, the methods +action_controller+, +action_view+ and +action_mailer+ aren't defined on the +Rails::Application::Configuration+ object, rather they are caught by +method_missing+ which does this:

<ruby>
  def method_missing(name, *args, &blk)
    if name.to_s =~ config_key_regexp
      return $2 == '=' ? @options[$1] = args.first : @options[$1]
    end

    super
  end
</ruby>

Whilst this code is not obvious at first, a little bit of further explanation will help you understand. +config_key_regexp+ is another method (a private one, like +method_missing+) defined here:

<ruby>
  def config_key_regexp
    bits = config_keys.map { |n| Regexp.escape(n.to_s) }.join('|')
    /^(#{bits})(?:=)?$/
  end
</ruby>

As is +config_keys+:

<ruby>
  def config_keys
    ([ :active_support, :action_view ] +
      Railtie.plugin_names).map { |n| n.to_s }.uniq
  end
</ruby>

Aha! There we've got mention of +action_view+, but what is in +Railtie.plugin_names+? Most likely in this case the other frameworks.

h5. +Railtie.plugin_names+

I'm going to show you two methods since the third one, +self.plugin_name+, calls the second one, +self.plugins+ and they're right after each other:

<ruby>
  module Rails
    class Railtie
      def self.inherited(klass)
        @plugins ||= []
        @plugins << klass unless klass == Plugin
      end

      def self.plugins
        @plugins
      end

      def self.plugin_names
        plugins.map { |p| p.plugin_name }
      end
    end
  end
</ruby>

In here we see that we get the +plugin_names+ from a variable called +@plugins+... which we haven't seen yet. Through the power of the wonderful +inherited+ the +@plugins+ variable is populated. +inherited+ is called when a class inherits, or subclasses, from this class. Therefore we can determine that the other classes are probably inheriting or subclassing from +Rails::Railtie+.

h3. Serving a Request

Now that your application is fully initialized, it's now ready to start serving requests.

h4. _rails server_

For servers running through _rails server_ you may recall that this uses +Rails::Server+ which is a subclass of +Rack::Server+. Previously we covered the initialization process of Rack but not completely up to the point where the server was running. Now that's what we'll do. Back when the +Rack::Server+ class was first covered there was a mention of the +start+ method which we only touched on. It goes a little like this:

<ruby>
  def start
    if options[:debug]
      $DEBUG = true
      require 'pp'
      p options[:server]
      pp wrapped_app
      pp app
    end

    if options[:warn]
      $-w = true
    end

    if includes = options[:include]
      $LOAD_PATH.unshift *includes
    end

    if library = options[:require]
      require library
    end

    daemonize_app if options[:daemonize]
    write_pid if options[:pid]
    server.run wrapped_app, options
  end
</ruby>

We were at the point of explaining what +wrapped_app+ was before we dived into the Rails initialization process.Now that we have a +wrapped_app+ we pass it as the first argument to +server.run+. +server+ in this instance is defined like this:

<ruby>
  def server
    @_server ||= Rack::Handler.get(options[:server]) || Rack::Handler.default
  end
</ruby>

Our +options+ Hash is still the default, and there is no +server+ key set in +default_options+, so it will default to +Rack::Handler.default+. This code works like this:

<ruby>
  def self.default(options = {})
    # Guess.
    if ENV.include?("PHP_FCGI_CHILDREN")
      # We already speak FastCGI
      options.delete :File
      options.delete :Port

      Rack::Handler::FastCGI
    elsif ENV.include?("REQUEST_METHOD")
      Rack::Handler::CGI
    else
      begin
        Rack::Handler::Mongrel
      rescue LoadError => e
        Rack::Handler::WEBrick
      end
    end
  end
</ruby>


We don't have +PHP_FCGI_CHILDREN+ in our +ENV+, so it's not going to be +FastCGI+. We also don't have +REQUEST_METHOD+ in there, so it's not going to be +CGI+. If we have Mongrel installed it'll default to that and then finally it'll use WEBrick. For this, we'll assume a bare-bones installation and assume WEBrick. So from this we can determine our default handler is +Rack::Handler::WEBrick+.

(side-note: Mongrel doesn't install on 1.9. TODO: How do we format these anyway?)

h5. +Rack::Handler::WEBrick+

This class is subclassed from +WEBrick::HTTPServlet::AbstractServlet+ which is a class that comes with the Ruby standard library. This is the magical class that serves the requests and deals with the comings (requests) and goings (responses) for your server.


+Rack::Server+ has handlers for the request and by default the handler for a _rails server_ server is

h3. Cruft!

The final line of _config/environment.rb_:

<ruby>
  YourApp::Application.initialize!
</ruby>

gets down to actually initializing the application!

TODO: Cover the other +config.*+ methods in perhaps a "Bonus" section near the end. If they aren't referenced in a config file they aren't that important, right?


TODO: This belongs in the guide, I just don't know where yet. Maybe towards the end, since this is really the "final" thing to be done before being able to serve requests.

<ruby>
  def build_app(app)
    middleware[options[:environment]].reverse_each do |middleware|
      middleware = middleware.call(self) if middleware.respond_to?(:call)
      next unless middleware
      klass = middleware.shift
      app = klass.new(app, *middleware)
    end
    app
  end
</ruby>

Because we don't have any middleware for our application, this returns the application itself( Guessing here!! TODO: Investigate if this is really the case.)

Now that we have an app instance, the last line in +start+ calls +server.run wrapped_app, options+. We know what our app is, and that our options are just the default options, so what is +server+? +server+ is this:

<ruby>
  def server
    @_server ||= Rack::Handler.get(options[:server]) || Rack::Handler.default
  end
</ruby>

Since we have default options, the server is obviously going to be +Rack::Handler.default+. The +default+ method goes like this:

<ruby>
  def self.default(options = {})
    # Guess.
    if ENV.include?("PHP_FCGI_CHILDREN")
      # We already speak FastCGI
      options.delete :File
      options.delete :Port

      Rack::Handler::FastCGI
    elsif ENV.include?("REQUEST_METHOD")
      Rack::Handler::CGI
    else
      begin
        Rack::Handler::Mongrel
      rescue LoadError => e
        Rack::Handler::WEBrick
      end
    end
  end
</ruby>

h3. +Rails::Paths+


The +super+ method it references comes from +Rails::Engine::Configuration+ which defines these paths:

<ruby>
  def paths
    @paths ||= begin
      paths = Rails::Paths::Root.new(@root)
      paths.app                 "app",                 :eager_load => true, :glob => "*"
      paths.app.controllers     "app/controllers",     :eager_load => true
      paths.app.helpers         "app/helpers",         :eager_load => true
      paths.app.models          "app/models",          :eager_load => true
      paths.app.views           "app/views"
      paths.lib                 "lib",                 :load_path => true
      paths.lib.tasks           "lib/tasks",           :glob => "**/*.rake"
      paths.lib.templates       "lib/templates"
      paths.config              "config"
      paths.config.initializers "config/initializers", :glob => "**/*.rb"
      paths.config.locales      "config/locales",      :glob => "*.{rb,yml}"
      paths.config.routes       "config/routes.rb"
      paths
    end
  end
</ruby>

h3. Appendix A

This file is _activesupport/lib/active_support/inflector/inflections.rb_ and defines the +ActiveSupport::Inflector::Inflections+ class which defines the +singularize+, +pluralize+, +humanize+, +tableize+, +titleize+ and +classify+ methods as well as the code to defining how to work out the irregular, singular, plural and human versions of words. These methods are called +irregular+, +singular+, +plural+ and +human+ respectively, as is the Rails way.

This file is _activesupport/lib/active_support/inflector/transliterate.rb_ and defines two methods, +transliterate+ and +parameterize+.

This file first requires _activesupport/lib/active_support/core_ext/string/multibyte.rb_, which requires _activesupport/lib/active_support/multibyte.rb_, which subsequently requires _activesupport/core_ext/module/attribute_accessors.rb_. The _attribute_accessors.rb_ file is needed to gain access to the +mattr_accessor+ (module attribute accessor) method, which is called in _active_suport/multibyte.rb_. The file _active_support/multibyte.rb_ also autoloads three other classes:

<ruby>
module ActiveSupport #:nodoc:
  module Multibyte
    autoload :EncodingError, 'active_support/multibyte/exceptions'
    autoload :Chars, 'active_support/multibyte/chars'
    autoload :Unicode, 'active_support/multibyte/unicode'
    ...
  end
end
</ruby>

There are also these method definitions:

<ruby>
  # The proxy class returned when calling mb_chars. You can use this accessor to configure your own proxy
  # class so you can support other encodings. See the ActiveSupport::Multibyte::Chars implementation for
  # an example how to do this.
  #
  # Example:
  #   ActiveSupport::Multibyte.proxy_class = CharsForUTF32
  def self.proxy_class=(klass)
    @proxy_class = klass
  end

  # Returns the currect proxy class
  def self.proxy_class
    @proxy_class ||= ActiveSupport::Multibyte::Chars
  end
</ruby>

These methods are used in _activesupport/lib/active_support/core_ext/string/multibyte.rb_.

The file _activesupport/lib/active_support/core_ext/string/chars.rb_  defines the default proxy class that will be returned by +mb_chars+.

Because Ruby 1.9's +String+ class has support for multibyte encodings, some methods are defined only for Ruby 1.8:

* +self.wants?+
* +++
* +=~+
* +=~+
* +center+
* +include?+
* +index+
* +insert+
* +ljust+
* +lstrip+, +lstrip!+
* +ord+
* +rindex+
* +rjust+
* +rstrip+, +rstrip!+
* +size+
* +strip+, +strip!+

However, Ruby 1.9 lacks support for some needed operations, so the following methods are defined for both Ruby 1.8 and Ruby 1.9:

* +<=>+
* +[]=+
* +capitalize+, +capitalize!+
* +compose+
* +decompose+
* +downcase+, +downcase!+
* +g_length+
* +limit+
* +normalize+
* +reverse+, +reverse+!
* +slice+, +slice!+
* +split+
* +tidy_bytes+, +tidy_bytes!+
* +titleize+
* +upcase+, +upcase!+

<ruby>
  class String
    if RUBY_VERSION >= "1.9"
      def mb_chars
        if ActiveSupport::Multibyte.proxy_class.consumes?(self)
          ActiveSupport::Multibyte.proxy_class.new(self)
        else
          self
        end
      end

      def is_utf8? #:nodoc
        case encoding
        when Encoding::UTF_8
          valid_encoding?
        when Encoding::ASCII_8BIT, Encoding::US_ASCII
          dup.force_encoding(Encoding::UTF_8).valid_encoding?
        else
          false
        end
      end
    else
      def mb_chars
        if ActiveSupport::Multibyte.proxy_class.wants?(self)
          ActiveSupport::Multibyte.proxy_class.new(self)
        else
          self
        end
      end

      # Returns true if the string has UTF-8 semantics (a String used for purely byte resources is unlikely to have
      # them), returns false otherwise.
      def is_utf8?
        ActiveSupport::Multibyte::Chars.consumes?(self)
      end
    end
</ruby>

As you can see, +mb_chars+ is where the +proxy_class+ property comes in handy. This method will create a new instance of the configured proxy class using the instance of +String+ as a constructor argument. By default, the new +String+-like object will be an instance of the proxy class +ActiveSupport::Multibyte::Chars+. You can use +ActiveSupport::Multibyte.proxy_class=+ to set a different proxy class if you wish.

Here, +mb_chars+ invokes +is_utf8?+ to checks if the string can be treated as UTF-8. On 1.9, the string's +encoding+ property is checked. On 1.8, +wants?+ checks to see if +$KCODE+ is "UTF-8" and, and +consumes?+ checks whether the string can be unpacked as UTF-8 without raising an error.

The keen eye will have seen +ActiveSupport::Multibyte::Chars+ was specified as an +autoload+ earlier: _activesupport/lib/active_support/multibyte/chars.rb_ will be loaded without an explicit +require+ when we call +is_utf8+ on 1.8, or +mb_chars+ on any Ruby version. This file includes _activesupport/lib/active_support/string/access.rb_ which defines methods such as +at+, +from+, +to+, +first+ and +last+. These methods will return parts of the string depending on what is passed to them.

The second file included is _activesupport/lib/active_support/string/behavior.rb_ which only defines  +acts_like_string?+ on +String+, a method which always returns +true+. This method is used by +Object#acts_like?+, which accepts a single argument representing the downcased and symbolised version of a class, and returns true if the object's behavior is like that class. In this case the code would be +acts_like?(:string)+.

The +Chars+ class also defines other important methods such as the "spaceship" method +<=>+, which is needed by the +Comparable+ module, in order to allow UTF-8-aware sorting.

h3. Common Includes

TODO: I feel this section would be better at the end of the guide as it breaks the flow.

This section is for all the common includes in the Railties.

h4. +require 'active_support/inflector'+

This file is _activesupport/lib/active_support/inflector.rb_ and makes a couple of requires out different files tasked with putting inflections in place:

<ruby>
  require 'active_support/inflector/inflections'
  require 'active_support/inflector/transliterate'
  require 'active_support/inflector/methods'

  require 'active_support/inflections'
  require 'active_support/core_ext/string/inflections'
</ruby>

The files included here define methods for modifying strings, such as +transliterate+ which will convert a Unicode string to its ASCII version, +parameterize+ for making strings into url-safe versions, +camelize+ for camel-casing a string such as +string_other+ into +StringOther+ and +ordinalize+ converting a string such as +101+ into +101st+. More information about these methods can be found in the Active Support Core Extensions Guide. TODO: Link to AS Guide.

h4. +require 'active_support/core_ext/module/delegation'+

_activesupport/lib/active_support/core_ext/module/delegation.rb_ defines the +delegate+ method which can be used to delegate methods to other methods in your code. Take the following code example:

<ruby>
  class Client < ActiveRecord::Base
    has_one :address

    delegate :address_line_1, :to => :address
  end
</ruby>

This defines an +address_line_1+ method which is defined as:

<ruby>
  def address_line_1(*args, &block)
    address.__send__(:address_line_1, *args, &block)
    rescue NoMethodError
      if address.nil?
        raise "address_line_1 is delegated to address.address_line_1, but address is nil: #{client.inspect}"
      end
  end
</ruby>

h4. +require 'active_support/core_ext/class/attribute_accessors'+

The file, _activesupport/lib/active_support/core_ext/class/attribute_accessors.rb_, defines the class accessor methods +cattr_writer+, +cattr_reader+ and +cattr_accessor+. +cattr_accessor+ defines a +cattr_reader+ and +cattr_writer+ for the symbol passed in. These methods work by defining class variables when you call their dynamic methods.

Throughout the Railties there a couple of common includes. They are listed here for your convenience.

h4. +require 'active_support/core_ext/module/attr_internal+

This file defines three methods +attr_internal_reader+, +attr_internal_writer+ and +attr_internal_accessor+. These work very similar to the +attr_reader+, +attr_writer+ and +attr_accessor+ methods, except the variables they define begin with +@_+. This was done to ensure that they do not clash with variables used people using Rails, as people are less-likely to define say, +@_request+ than they are to define +@request+. An example of where this method is used is for +params+ in the +ActionController::Metal+ class.

h4. +require 'active_support/ruby/shim'+

The _activesupport/lib/active_support/ruby/shim.rb_ file requires methods that have been implemented in Ruby versions greater than 1.9. This is done so you can use Rails 3 on versions earlier than 1.9, such as 1.8.7. These methods are:

* +Date#next_month+
* +Date#next_year+
* +DateTime#to_date+
* +DateTime#to_datetime+
* +DateTime#xmlschema+
* +Enumerable#group_by+
* +Enumerable#each_with_object+
* +Enumerable#none?+
* +Process#daemon+
* +String#ord+
* +Time#to_date+
* +Time.to_time+
* +Time.to_datetime+

For more information see the Active Support Core Extensions guide TODO: link to relevant sections for each method.

And "the REXML security fix detailed here":[http://weblog.rubyonrails.org/2008/8/23/dos-vulnerabilities-in-rexml]