basic rails sitemap generator

routes.rb

  get '/sitemap', :to => 'sitemap#sitemap', :as => 'sitemap'

Sitemap.rb

# -*- encoding : utf-8 -*-
class Sitemap
  def self.static_view_templates
    require 'fileutils'
    template_paths = []
    FileUtils.cd Rails.root.join 'app', 'views', 'static_pages' do
      erb_html_files = File.join "**", "*.html.erb"
      template_paths = Dir.glob(erb_html_files)
    end
    remove_html_erb!(template_paths)
    template_paths -= ['home']
    template_paths += ['', 'members/join']
    template_paths
  end

private
  def self.remove_html_erb!(template_paths)
    template_paths.each {|p| p.sub!(/\.html\.erb/, '')}
  end
end

sitemap_controller.rb

# -*- encoding : utf-8 -*-
class SitemapController < ApplicationController
  respond_to :xml

  def sitemap
    response.headers['Cache-Control'] = "public, max-age=#{1.week}"
    @page_paths = Sitemap.static_view_templates
  end
end

sitemap_controller_spec.rb

# -*- encoding : utf-8 -*-
require 'spec_helper'

describe SitemapController do
  describe "GET sitemap(.xml)" do
    subject {get :sitemap, :format => 'xml'}

    describe "static pages" do
      # create some test templates
      around(:each) do |example|
        paths = [
          Rails.root.join('app', 'views', 'static_pages', 'testpages-50-top-ukcat-tips', '01-what-is-ukcat-and-is-it-fair.html.erb'),
          Rails.root.join('app', 'views', 'static_pages', 'testpages-privacy.html.erb')
        ]
        paths.each {|path| `mkdir -p #{File.dirname(path)} && touch #{path}` }
        example.run
        paths.each {|path| `rm #{path}; rmdir #{File.dirname(path)}` }
      end
      
      it "should assign the static page paths list to @page_paths" do
        # check some of your static pages here, if any
        subject
        assigns[:page_paths].should include('testpages-privacy')
        assigns[:page_paths].should include('testpages-50-top-ukcat-tips/01-what-is-ukcat-and-is-it-fair')
      end
    end
    
    it "should not include home, but should include the root path" do
      # you may or may not need this test, depends on your static page controller or cms (if you have one)
      subject
      assigns[:page_paths].should include('')
      assigns[:page_paths].should_not include('home')
      assigns[:page_paths].should_not include('/home')
      assigns[:page_paths].should_not include('/home/')
    end

    it "should include the member signup page" do
      # business critical, or check for all your public facing pages being included; ours are mostly paywalled
      subject
      assigns[:page_paths].should include('members/join')
    end

    it {should render_template('sitemap/sitemap')}

    it "should respond with xml" do
      subject
      response.content_type.should == 'application/xml'
      response.charset.should == 'utf-8'
    end
  end
end

sitemap_request_spec.rb

# -*- encoding : utf-8 -*-
require 'spec_helper'

describe "/sitemap" do
# Feature: Sitemap
#   In order to get siterank
#   As a site owner
#   I want a sitemap generated for our static pages

  # create some test templates
  around(:each) do |example|
    paths = [
      Rails.root.join('app', 'views', 'static_pages', 'testpages-50-top-ukcat-tips', '01-what-is-ukcat-and-is-it-fair.html.erb'),
      Rails.root.join('app', 'views', 'static_pages', 'testpages-privacy.html.erb')
    ]
    paths.each {|path| `mkdir -p #{File.dirname(path)} && touch #{path}` }
    example.run
    paths.each {|path| `rm #{path}; rmdir #{File.dirname(path)}` }
  end

  specify "the sitemap is served correctly and includes /, home, a level 1 and a nested page" do
    visit '/sitemap.xml'
    response_content_type_should_be "application/xml; charset=utf-8"
    page_source_should_match %Q{<?xml version="1.0" encoding="UTF-8"?>}
    within 'urlset[xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"]' do
      check_link_present_with_weekly_changefreq '/'
      check_link_present_with_weekly_changefreq '/members/join'
      check_link_present_with_weekly_changefreq '/testpages-privacy'
      check_link_present_with_weekly_changefreq '/testpages-50-top-ukcat-tips/01-what-is-ukcat-and-is-it-fair'
      check_all_static_templates_linked
    end
    response_body_should_satisfy_xml_schema "http://www.sitemaps.org/schemas/sitemap/0.9/sitemap.xsd"
  end
  
  def check_link_present_with_weekly_changefreq(path)
    page.should have_xpath("//url/loc[.='http://www.example.com#{path}']/../changefreq[.='weekly']")    
  end
  
  def check_all_static_templates_linked
    require 'fileutils'
    template_paths = []
    FileUtils.cd Rails.root.join 'app', 'views', 'static_pages' do
      erb_html_files = File.join "**", "*.html.erb"
      template_paths = Dir.glob(erb_html_files)
    end
    template_paths = template_paths.map {|p| p.sub(/\.html\.erb/, '')}
    template_paths = template_paths - ['home']
    template_paths.each do |path|
      check_link_present_with_weekly_changefreq('/' + path)
    end
  end
end