Skip to content
This repository
tree: 7c79996c67
Fetching contributors…

Octocat-spinner-32-eaf2f5

Cannot retrieve contributors at this time

file 186 lines (167 sloc) 7.305 kb
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185
require 'set'

module ActionController #:nodoc:
  module Caching
    # Action caching is similar to page caching by the fact that the entire
    # output of the response is cached, but unlike page caching, every
    # request still goes through Action Pack. The key benefit of this is
    # that filters run before the cache is served, which allows for
    # authentication and other restrictions on whether someone is allowed
    # to execute such action. Example:
    #
    # class ListsController < ApplicationController
    # before_filter :authenticate, :except => :public
    #
    # caches_page :public
    # caches_action :index, :show
    # end
    #
    # In this example, the +public+ action doesn't require authentication
    # so it's possible to use the faster page caching. On the other hand
    # +index+ and +show+ require authentication. They can still be cached,
    # but we need action caching for them.
    #
    # Action caching uses fragment caching internally and an around
    # filter to do the job. The fragment cache is named according to
    # the host and path of the request. A page that is accessed at
    # <tt>http://david.example.com/lists/show/1</tt> will result in a fragment named
    # <tt>david.example.com/lists/show/1</tt>. This allows the cacher to
    # differentiate between <tt>david.example.com/lists/</tt> and
    # <tt>jamis.example.com/lists/</tt> -- which is a helpful way of assisting
    # the subdomain-as-account-key pattern.
    #
    # Different representations of the same resource, e.g.
    # <tt>http://david.example.com/lists</tt> and
    # <tt>http://david.example.com/lists.xml</tt>
    # are treated like separate requests and so are cached separately.
    # Keep in mind when expiring an action cache that
    # <tt>:action => 'lists'</tt> is not the same as
    # <tt>:action => 'list', :format => :xml</tt>.
    #
    # You can modify the default action cache path by passing a
    # <tt>:cache_path</tt> option. This will be passed directly to
    # <tt>ActionCachePath.path_for</tt>. This is handy for actions with
    # multiple possible routes that should be cached differently. If a
    # block is given, it is called with the current controller instance.
    #
    # And you can also use <tt>:if</tt> (or <tt>:unless</tt>) to pass a
    # proc that specifies when the action should be cached.
    #
    # Finally, if you are using memcached, you can also pass <tt>:expires_in</tt>.
    #
    # The following example depicts some of the points made above:
    #
    # class ListsController < ApplicationController
    # before_filter :authenticate, :except => :public
    #
    # caches_page :public
    #
    # caches_action :index, :if => proc do
    # !request.format.json? # cache if is not a JSON request
    # end
    #
    # caches_action :show, :cache_path => { :project => 1 },
    # :expires_in => 1.hour
    #
    # caches_action :feed, :cache_path => proc do
    # if params[:user_id]
    # user_list_url(params[:user_id, params[:id])
    # else
    # list_url(params[:id])
    # end
    # end
    # end
    #
    # If you pass <tt>:layout => false</tt>, it will only cache your action
    # content. That's useful when your layout has dynamic information.
    #
    # Warning: If the format of the request is determined by the Accept HTTP
    # header the Content-Type of the cached response could be wrong because
    # no information about the MIME type is stored in the cache key. So, if
    # you first ask for MIME type M in the Accept header, a cache entry is
    # created, and then perform a second request to the same resource asking
    # for a different MIME type, you'd get the content cached for M.
    #
    # The <tt>:format</tt> parameter is taken into account though. The safest
    # way to cache by MIME type is to pass the format in the route.
    module Actions
      extend ActiveSupport::Concern

      module ClassMethods
        # Declares that +actions+ should be cached.
        # See ActionController::Caching::Actions for details.
        def caches_action(*actions)
          return unless cache_configured?
          options = actions.extract_options!
          options[:layout] = true unless options.key?(:layout)
          filter_options = options.extract!(:if, :unless).merge(:only => actions)
          cache_options = options.extract!(:layout, :cache_path).merge(:store_options => options)

          around_filter ActionCacheFilter.new(cache_options), filter_options
        end
      end

      def _save_fragment(name, options)
        content = ""
        response_body.each do |parts|
          content << parts
        end

        if caching_allowed?
          write_fragment(name, content, options)
        else
          content
        end
      end

    protected
      def expire_action(options = {})
        return unless cache_configured?

        if options.is_a?(Hash) && options[:action].is_a?(Array)
          options[:action].each {|action| expire_action(options.merge(:action => action)) }
        else
          expire_fragment(ActionCachePath.new(self, options, false).path)
        end
      end

      class ActionCacheFilter #:nodoc:
        def initialize(options, &block)
          @cache_path, @store_options, @cache_layout =
            options.values_at(:cache_path, :store_options, :layout)
        end

        def filter(controller)
          path_options = if @cache_path.respond_to?(:call)
            controller.instance_exec(controller, &@cache_path)
          else
            @cache_path
          end

          cache_path = ActionCachePath.new(controller, path_options || {})

          body = controller.read_fragment(cache_path.path, @store_options)

          unless body
            controller.action_has_layout = false unless @cache_layout
            yield
            controller.action_has_layout = true
            body = controller._save_fragment(cache_path.path, @store_options)
          end

          body = controller.render_to_string(:text => body, :layout => true) unless @cache_layout

          controller.response_body = body
          controller.content_type = Mime[cache_path.extension || :html]
        end
      end

      class ActionCachePath
        attr_reader :path, :extension

        # If +infer_extension+ is true, the cache path extension is looked up from the request's
        # path and format. This is desirable when reading and writing the cache, but not when
        # expiring the cache - expire_action should expire the same files regardless of the
        # request format.
        def initialize(controller, options = {}, infer_extension = true)
          if infer_extension
            @extension = controller.params[:format]
            options.reverse_merge!(:format => @extension) if options.is_a?(Hash)
          end

          path = controller.url_for(options).split(%r{://}).last
          @path = normalize!(path)
        end

      private
        def normalize!(path)
          path << 'index' if path[-1] == ?/
          path << ".#{extension}" if extension and !path.split('?').first.ends_with?(".#{extension}")
          URI.parser.unescape(path)
        end
      end
    end
  end
end
Something went wrong with that request. Please try again.