Skip to content
Commits on Oct 2, 2010
  1. @josevalim

    Deprecate config.generators in Rails::Engine in favor of config.app_g…

    josevalim committed Oct 2, 2010
    config.generators still works fine in Rails::Application and Rails::Railtie, but it will be deprecated in the latter in Rails 3.1.
Commits on Aug 15, 2010
  1. @wycats
Commits on Jun 27, 2010
  1. @josevalim

    A few changes were done in this commit:

    josevalim committed Jun 28, 2010
    * Added :autoload to engines path API and redefine usage to be in sync with 6f83a50;
    * Do not autoload code in *lib* for applications (now you need to explicitly require them). This makes an application behave closer to an engine (code in lib is still autoloaded for plugins);
    * Always autoload code in app/ for engines and plugins. This makes engines behave closer to an application and should allow us to get rid of the unloadable hack required when controllers inside engines inherit from ApplicationController;
Commits on Jun 24, 2010
  1. @josevalim

    Move Rails::LogSubscriber to ActiveSupport::LogSubscriber, allowing f…

    josevalim committed Jun 24, 2010
    …rameworks like ActiveRecord and ActiveResource to log outsude Rails::Application [#4816 state:resolved]
Commits on May 29, 2010
  1. @josevalim
  2. @wycats

    Removing Metal from Rails 3.

    wycats committed May 29, 2010
    If you have existing Metals, you have a few options:
    * if your metal behaves like a middleware, add it to the
      middleware stack via config.middleware.use. You can use
      methods on the middleware stack to control exactly where
      it should go
    * if it behaves like a Rack endpoint, you can link to it
      in the router. This will result in more optimal routing
      time, and allows you to remove code in your endpoint
      that matches specific URLs in favor of the more powerful
      handling in the router itself.
    For the future, you can use ActionController::Metal to get
    a very fast controller with the ability to opt-in to specific
    controller features without paying the penalty of the full
    controller stack.
    Since Rails 3 is closer to Rack, the Metal abstraction is
    no longer needed.
Commits on May 18, 2010
  1. @josevalim
Commits on May 17, 2010
  1. @wycats

    Revert "Moved encoding work in progress to a feature branch."

    wycats committed May 17, 2010
    This reverts commit ade756f.
Commits on May 16, 2010
  1. @jeremy

    Moved encoding work in progress to a feature branch.

    jeremy committed May 16, 2010
    This reverts commits af0d1a8 and 64d109e.
  2. @wycats

    Significantly improved internal encoding heuristics and support.

    wycats committed May 16, 2010
    * Default Encoding.default_internal to UTF-8
    * Eliminated the use of file-wide magic comments to coerce code evaluated inside the file
    * Read templates as BINARY, use default_external or template-wide magic comments
      inside the Template to set the initial encoding
      * This means that template handlers in Ruby 1.9 will receive Strings encoded
        in default_internal (UTF-8 by default)
    * Create a better Exception for encoding issues, and use it when the template
      source has bytes that are not compatible with the specified encoding
    * Allow template handlers to opt-into handling BINARY. If they do so, they
      need to do some of their own manual encoding work
    * Added a "Configuration Gotchas" section to the intro Rails Guide instructing
      users to use UTF-8 for everything
    * Use config.encoding= in Ruby 1.8, and raise if a value that is an invalid
      $KCODE value is used
    * Fixed a few tests that were assert() rather than assert_equal() and
      were caught by Minitest requiring a String for the message
    * Fixed a test where an assert_select was misformed, also caught by
      Minitest being more restrictive
    * Fixed a test where a Rack response was returning a String rather
      than an Enumerable
  3. @josevalim
Commits on May 15, 2010
  1. @wycats

    Reorganized initializers a bit to enable better hooks for common case…

    wycats committed May 15, 2010
    …s without the need for Railtie. Specifically, the following hooks were added:
    * before_configuration: this hook is run immediately after the Application class 
      comes into existence, but before the user has added any configuration. This is
      the appropriate place to set configuration for your plugin
    * before_initialize: This is run after all of the user's configuration has completed,
      but before any initializers have begun (in other words, it runs right after
    * after_initialize: This is run after all of the initializers have run. It is an
      appropriate place for forking in a preforking setup
    Each of these hooks may be used via ActiveSupport.on_load(name) { }. In all these cases, the context inside the block will be the Application object. This means that for simple cases, you can use these hooks without needing to create a Railtie.
Commits on Apr 29, 2010
  1. @jeremy

    Add missing deprecation requires

    jeremy committed Apr 29, 2010
  2. @josevalim
  3. @josevalim

    The rake task :environment now loads config/environment.rb instead of…

    josevalim committed Apr 29, 2010
    … initializing the application on its own. This fixes [#4492 state:resolved] and also avoids the application being initialized twice in some rake tasks.
Commits on Apr 22, 2010
  1. @josevalim

    Clean up the config object in ActionPack. Create config_accessor whic…

    josevalim committed Apr 22, 2010
    …h just delegates to the config object, reducing the number of deprecations and add specific tests.
Commits on Apr 12, 2010
  1. @anildigital
Commits on Apr 8, 2010
  1. @josevalim

    Fix a bug in ActionDispatch::Static where Rails cannot find assets if…

    josevalim committed Apr 8, 2010
    … started in another directory which is not the RAILS_ROOT.
Commits on Apr 7, 2010
  1. @josevalim
Commits on Apr 5, 2010
  1. @josevalim

    Rename config.cookie_secret to config.secret_token and pass it as con…

    josevalim committed Apr 5, 2010
    …figuration in request.env. This is another step forward removing global configuration.
Commits on Apr 2, 2010
  1. @rolfb @josevalim

    Raise exceptions instead of rendering error templates in test environ…

    rolfb committed with josevalim Apr 2, 2010
    …ment [#4315 state:resolved]
    Signed-off-by: José Valim <>
Commits on Mar 26, 2010
  1. @josevalim

    Move application configuration to the application configuration objec…

    josevalim committed Mar 26, 2010
    …t, remove railtie_name and engine_name and allow to set the configuration object.
Commits on Mar 20, 2010
  1. @josh

    Move railties/builtin into lib

    josh committed Mar 20, 2010
Commits on Mar 8, 2010
  1. Move middlewares to the Application level. If you want to modify the …

    Carlhuda committed Mar 8, 2010
    …middleware, either use Rails.application.config.middleware or modify it in an initializer (if you need to set it relative to user-specified middleware).
Commits on Mar 2, 2010
  1. Have log subscribers subscribe to the actual events, so the subscribe…

    Carlhuda committed Mar 2, 2010
    …r doesn't subscribe to *every* event, so we can have events that are slow-ish but are not actually run in production.
Commits on Feb 1, 2010
  1. @josevalim
Commits on Jan 28, 2010
  1. @josevalim

    Add reloadable specific for engines and move environment to applicati…

    josevalim committed with Carl Lerche Jan 27, 2010
    …on paths.
    Signed-off-by: Carl Lerche <>
Commits on Jan 23, 2010
  1. @josevalim
Something went wrong with that request. Please try again.