Skip to content

newsdeeply/wp-connector

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

wp-connector

Scrutinizer Code Quality Build Status Code Climate Test Coverage

This library is part of the WP on Rails project, that uses WordPress (WP) in a "headless" fashion —as a means of managing content— while using a Rails application to serve public requests and provide a basis for customizations.

To prepare a WP installation to be used in a WP on Rails architecture the wp-relinquish plugin is provided. It provides a way to configure WP actions to send notifications as webhook calls to the Rails application equipped with wp-connector.

When using WP on Rails the content's master data resides in WP's database, as that's where is it created and modified. The Rails application that is connected to WP stores a structured copy of the data (usually providing a separate table for each datatype), acting as a cache, on the basis of which the public requests are served.

The main reasons for not using WP to serve public web requests are:

  • Security — The internet is a dangerous place and WordPress has proven to be a popular target for malicious hackers. By not serving public requests from WP, but only the admin interface, the attack surface is significantly reduced.
  • Performance — Performance tuning WP can be difficult, especially when a generic caching-proxy (such as Varnish) is not viable due to dynamic content such as ads or personalization. Application frameworks provide a means for fine-grained caching strategies that are needed to serve high-traffic websites containing dynamic content.
  • Cost (TCO) of customizations — Customizing WP, and maintaining those customizations, is costly (laborious) and risky (error prone) compared to building custom functionality on top of an application framework (which is specifically designed for that purpose).
  • Upgrade path — Keeping a customized WP installation up-to-date can be a pain, and WP-updates come ever more often. When WP is not used to serve public requests and customizations are not built into WP most of this pain avoided.
  • Modern browser interfaces — The new breed of JavaScript UI frameworks (such as Facebook's React and Google's AngularJS) have features like "isomorphism" (allowing JS views to be pre-rendered on the server) and "two-way data-binding." These features expect certain server-side behavior and usually consume JSON data instead of HTML. The Rails community provides many tools to facilitate these new UI libraries from the server-side.

How it works

After the Rails application receives the webhook call from WP, simply notifying that some content is created or modified, a delayed job to fetch the content is scheduled using Sidekiq. The content is not fetched immediately, but scheduled for a fraction of a second later, the reason for this is twofold:

  1. The webhook call is synchronous, responding as soon as possible is needed to keep the admin interface of WP responsive.
  2. It is not guaranteed that all WP's processing is complete (some actions may still fire) by the time the webhook call is made.

The delayed job fetches the relevant content from WP using the WP-REST-API (this can be one or more requests), then possibly transforms and/or enriches the data, and finally stores it using a regular ActiveRecord model. The logic for the fetch and transform/enrich steps is simply part of the ActiveRecord model definition.

Installation

Add this line to your Rails application's Gemfile:

gem 'wp-connector', :github => 'wponrails/wp-connector'

Then execute bundle install.

Usage

In WordPress install both the wp-relinquish and json-rest-api plugins. The wonderful ACF plugin should work out-of-the-box with wp-relinquish. The wp-relinquish plugin needs some configuration, as specified in the README of that project.

Create an initializer and specify the "JSON route" of the WP REST API, by setting the wordpress_url config option:

# config/initializers/wp_connector.rb

WpConnector.configure do |config|
  config.wordpress_url = "http://wordpress-site.dev/"
end

Here wordpress-site.dev is the domain for your Wordpress site.

In order to configure models, ['articles', 'news_articles', 'pages'], for paginated requests to the WP-REST API, add the following to your initializer:

config.wp_api_paginated_models = %w(articles news_articles pages)

Also specify the wp-connector API key in the wp-connector configuration by adding the wp_connector_api_key config option to the initializer:

config.wp_connector_api_key="H3O5P6P1I5N8G8E4R"

Create the routes for the webhook endpoint (in config/routes.rb of your Rails app):

# wp-connector endpoints
post   'wp-connector/:model',     to: 'wp_connector#model_save'
delete 'wp-connector/:model/:id', to: 'wp_connector#model_delete'

Create a WpConnectorController class (in app/controllers/wp_connector_controller.rb) that specifies a webhook action. For example for the Post type:

class WpConnectorController < ApplicationController
  include WpWebhookEndpoint
  skip_before_action :verify_authenticity_token

  def model_save
    model = params[:model].classify.constantize
    render_json_200_or_404 model.schedule_create_or_update(wp_id_from_params)
  end

  def model_delete
    model = params[:model].constantize
    render_json_200_or_404 model.purge(wp_id_from_params)
  end
end

Create a model for each of the content types that need to be cached by the Rails application. This is an example for the Post model:

class Post < ActiveRecord::Base
  include WpCache
  include WpPost

  def update_wp_cache(json)
    update_post(json)
    author = Author.find_or_create(json["author"]["ID"])
    author.update_wp_cache(json["author"])
    self.author = author
    self.save
  end
end

And an example of an Author model:

class Author < ActiveRecord::Base
  include WpCache
  include WpPost

  def update_wp_cache(json)
    update_post(json)
    self.username     = json["username"]
    self.name         = json["name"]
    self.first_name   = json["first_name"]
    self.last_name    = json["last_name"]
    self.nickname     = json["nickname"]
    self.url          = json["URL"]
    self.description  = json["description"]
    self.registered   = json["registered"]
    self.save
  end
end

And create the migration for these models:

class CreatePostsAndAuthors < ActiveRecord::Migration
  def change
    create_table :posts do |t|
      t.integer :wp_id
      t.string  :title
      t.integer :author_id
      t.text    :content
      t.string  :slug
      t.text    :excerpt
      t.timestamps
    end

    create_table :authors do |t|
      t.integer :wp_id
      t.string  :username
      t.string  :name
      t.string  :first_name
      t.string  :last_name
      t.string  :nickname
      t.string  :slug
      t.string  :url
      t.string  :description
      t.string  :registered
      t.timestamps
    end
  end
end

wp-connector assumes the model name is the same as the wp post type name. This means the API call for the author becomes:

http://wordpress-site.dev/?json_route=/authors/{id}

However, in some cases the model names do not correspond. Therefore the wp_type of a rails model can be overridden with the wp_type class method.

class Author < ActiveRecord::Base
  # ... snip ...

  def self.wp_type
    'wp_post_type_name'
  end

  # ... snap ...
end

Resulting in the following API call:

http://wordpress-site.dev/?json_route=/wp_post_type_name/{id}

Menu items

For menu items the json-rest-api-menu-routes WordPress plugin is required.

The implementation for menus is rather straightforward. An example for a basic Menu model:

class Menu < ActiveRecord::Base
  include WpCache
  include WpMenu

  def update_wp_cache(json)
    update_menu(json)
  end
end

And the migration would be:

class CreateMenus < ActiveRecord::Migration
  def change
    create_table :menus do |t|
      t.integer :wp_id
      t.string  :slug
      t.text    :description
      t.string  :slug
      t.integer :count
      t.text    :items
      t.timestamps
    end
  end
end

Contributing

You know the drill :)

  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. Submit a "Pull Request".

License

Copyright (c) 2014-2015, Hoppinger B.V.

All files in this repository are MIT-licensed, as specified in the LICENSE file.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Ruby 100.0%