actionpack-page_caching

Static page caching for Action Pack (removed from core in Rails 4.0).

Page caching is an approach to caching in which response bodies are stored in files that the web server can serve directly:

1. A request to endpoint E arrives.
2. Its response is calculated and stored in a file F.
3. Next time E is requested, the web server sends F directly.

That applies only to GET or HEAD requests whose response code is 200, the rest are ignored.

Unlike caching proxies or other more sophisticated setups, page caching results in a dramatic speed up while being dead simple at the same time. Awesome cost/benefit.

The reason for such performance boost is that cached endpoints are short-circuited by the web server, which is very efficient at serving static files. Requests to cached endpoints do not even reach your Rails application.

This technique, however, is only suitable for pages that do not need to go through your Rails stack, precisely. For example, content management systems like wikis have typically many pages that are a great fit for this approach, but account-based systems where people log in and manipulate their own data are often less likely candidates. As a use case you can check, Rails Contributors makes heavy use of page caching. Its source code is here.

It is not all or nothing, though, in HTML cached pages JavaScript can still tweak details here and there dynamically as a trade-off.

Add this line to your application's Gemfile:

gem "actionpack-page_caching"

And then execute:

$ bundle

Page caching needs caching enabled:

config.action_controller.perform_caching = true

That goes typically in config/environments/production.rb, but you can activate that flag in any mode.

Since Rails 5 there is a special toggler to easily enable/disable caching in development mode without editing its configuration file. Just execute

$ bin/rails dev:cache

to enable/disable caching in development mode.

By default, files are stored below the public directory of your Rails application, with a path that matches the one in the URL.

For example, a page-cached request to /posts/what-is-new-in-rails-6 would be stored by default in the file public/posts/what-is-new-in-rails-6.html, and the web server would be configured to check that path in the file system before falling back to Rails. More on this later.

The default page caching directory can be overridden:

config.action_controller.page_cache_directory = Rails.root.join("public", "cached_pages")

There is no need to ensure the directory exists when the application boots, whenever a page has to be cached, the page cache directory is created if needed.

The globally configured cache directory, default or custom, can be overridden in each controller. There are three ways of doing this.

With a lambda:

class WeblogController < ApplicationController
  self.page_cache_directory = -> {
    Rails.root.join("public", request.domain)
  }
end

a symbol:

class WeblogController < ApplicationController
  self.page_cache_directory = :domain_cache_directory

  private
    def domain_cache_directory
      Rails.root.join("public", request.domain)
    end
end

or a callable object:

class DomainCacheDirectory
  def self.call(request)
    Rails.root.join("public", request.domain)
  end
end

class WeblogController < ApplicationController
  self.page_cache_directory = DomainCacheDirectory
end

Intermediate directories are created as needed also in this case.

Specifying which actions have to be cached is done through the caches_page class method:

class WeblogController < ActionController::Base
  caches_page :show, :new
end

The wiki of the project has some examples of web server configuration.

Expiration of the cache is handled by deleting the cached files, which results in a lazy regeneration approach in which the content is stored again as cached endpoints are hit.

If the cache is stored in a separate directory like public/cached_pages, you can easily expire the whole thing by removing said directory.

Removing a directory recursively with something like rm -rf is unreliable because that operation is not atomic and can mess up with concurrent page cache generation.

In POSIX systems moving a file is atomic, so the recommended approach would be to move the directory first out of the way, and then recursively delete that one. Something like

#!/bin/bash

tmp=public/cached_pages-$(date +%s)
mv public/cached_pages $tmp
rm -rf $tmp

As noted before, the page cache directory is created if it does not exist, so moving the directory is enough to have a clean cache, no need to recreate.

The API for doing so mimics the options from url_for and friends:

class WeblogController < ActionController::Base
  def update
    List.update(params[:list][:id], params[:list])
    expire_page action: "show", id: params[:list][:id]
    redirect_to action: "show", id: params[:list][:id]
  end
end

Additionally, you can expire caches using Sweepers that act on changes in the model to determine when a cache is supposed to be expired.

1. Fork it.
2. Create your feature branch (git checkout -b my-new-feature).
3. Commit your changes (git commit -am 'Add some feature').
4. Push to the branch (git push origin my-new-feature).
5. Create a new Pull Request.