sitemap_generator 6.3.0 → 7.0.2

data/lib/sitemap_generator/link_set.rb

@@ -1,11 +1,13 @@
+# frozen_string_literal: true
+
 require 'builder'
 
 # A LinkSet provisions a bunch of links to sitemap files.  It also writes the index file
 # which lists all the sitemap files written.
 module SitemapGenerator
   class LinkSet
-    @@requires_finalization_opts = [:filename, :sitemaps_path, :sitemaps_host, :namer]
-    @@new_location_opts = [:filename, :sitemaps_path, :namer]
+    @@requires_finalization_opts = %i[filename sitemaps_path sitemaps_host namer]
+    @@new_location_opts = %i[filename sitemaps_path namer]
 
     attr_reader :default_host, :sitemaps_path, :filename, :create_index
     attr_accessor :include_root, :include_index, :adapter, :yield_sitemap, :max_sitemap_links
@@ -30,17 +32,17 @@ module SitemapGenerator
     # If you are calling +create+ more than once in your sitemap configuration file,
     # make sure that you set a different +sitemaps_path+ or +filename+ for each call otherwise
     # the sitemaps may be overwritten.
-    def create(opts={}, &block)
+    def create(opts = {}, &block) # rubocop:disable Metrics/AbcSize,Metrics/MethodLength
       reset!
       set_options(opts)
       if verbose
         start_time = Time.now
         puts "In '#{sitemap_index.location.public_path}':"
       end
-      interpreter.eval(:yield_sitemap => yield_sitemap?, &block)
-      finalize!
+      interpreter.eval(yield_sitemap: yield_sitemap?, &block)
+      finalize! if block_given?
       end_time = Time.now if verbose
-      output(sitemap_index.stats_summary(:time_taken => end_time - start_time)) if verbose
+      output(sitemap_index.stats_summary(time_taken: end_time - start_time)) if verbose
       self
     end
 
@@ -117,21 +119,19 @@ module SitemapGenerator
     #
     # Note: When adding a new option be sure to include it in `options_for_group()` if
     # the option should be inherited by groups.
-    def initialize(options={})
+    def initialize(options = {})
       @default_host, @sitemaps_host, @yield_sitemap, @sitemaps_path, @adapter, @verbose, @protect_index, @sitemap_index, @added_default_links, @created_group, @sitemap = nil
 
       options = SitemapGenerator::Utilities.reverse_merge(options,
-        :include_root => true,
-        :include_index => false,
-        :filename => :sitemap,
-        :search_engines => {
-          :google         => "http://www.google.com/webmasters/tools/ping?sitemap=%s"
-        },
-        :create_index => :auto,
-        :compress => true,
-        :max_sitemap_links => SitemapGenerator::MAX_SITEMAP_LINKS
+        include_root: true,
+        include_index: false,
+        filename: :sitemap,
+        search_engines: {},
+        create_index: :auto,
+        compress: true,
+        max_sitemap_links: SitemapGenerator::MAX_SITEMAP_LINKS
       )
-      options.each_pair { |k, v| instance_variable_set("@#{k}".to_sym, v) }
+      options.each_pair { |k, v| instance_variable_set(:"@#{k}", v) }
 
       # If an index is passed in, protect it from modification.
       # Sitemaps can be added to the index but nothing else can be changed.
@@ -146,9 +146,9 @@ module SitemapGenerator
     # link - string link e.g. '/merchant', '/article/1' or whatever.
     # options - see README.
     #   host - host for the link, defaults to your <tt>default_host</tt>.
-    def add(link, options={})
-      add_default_links if !@added_default_links
-      sitemap.add(link, SitemapGenerator::Utilities.reverse_merge(options, :host => @default_host))
+    def add(link, options = {})
+      add_default_links unless @added_default_links
+      sitemap.add(link, SitemapGenerator::Utilities.reverse_merge(options, host: @default_host))
     rescue SitemapGenerator::SitemapFullError
       finalize_sitemap!
       retry
@@ -163,8 +163,8 @@ module SitemapGenerator
     #
     # The `:host` option defaults to the value of `sitemaps_host` which is the host where your
     # sitemaps reside.  If no `sitemaps_host` is set, the `default_host` is used.
-    def add_to_index(link, options={})
-      sitemap_index.add(link, SitemapGenerator::Utilities.reverse_merge(options, :host => sitemaps_host))
+    def add_to_index(link, options = {})
+      sitemap_index.add(link, SitemapGenerator::Utilities.reverse_merge(options, host: sitemaps_host))
     end
 
     # Create a new group of sitemap files.
@@ -194,7 +194,7 @@ module SitemapGenerator
     # Options like <tt>:default_host</tt> can be used and it will only affect the links
     # within the group.  Links added outside of the group will revert to the previous
     # +default_host+.
-    def group(opts={}, &block)
+    def group(opts = {}, &block) # rubocop:disable Metrics/AbcSize,Metrics/CyclomaticComplexity,Metrics/MethodLength,Metrics/PerceivedComplexity
       @created_group = true
       original_opts = opts.dup
 
@@ -217,7 +217,7 @@ module SitemapGenerator
         @original_location = @sitemap.location.dup
         @sitemap.location.merge!(@group.sitemap_location)
         if block_given?
-          @group.interpreter.eval(:yield_sitemap => @yield_sitemap || SitemapGenerator.yield_sitemap?, &block)
+          @group.interpreter.eval(yield_sitemap: @yield_sitemap || SitemapGenerator.yield_sitemap?, &block)
           @group.finalize_sitemap!
           @sitemap.location.merge!(@original_location)
         end
@@ -236,7 +236,7 @@ module SitemapGenerator
         @group.send(:create_index=, true, true) if @group.create_index != false
 
         if block_given?
-          @group.interpreter.eval(:yield_sitemap => @yield_sitemap || SitemapGenerator.yield_sitemap?, &block)
+          @group.interpreter.eval(yield_sitemap: @yield_sitemap || SitemapGenerator.yield_sitemap?, &block)
           @group.finalize_sitemap!
         end
       end
@@ -278,28 +278,28 @@ module SitemapGenerator
     # Is equivalent to:
     #
     #   SitemapGenerator::Sitemap.ping_search_engines('http://example.com/sitemap.xml.gz', :super_engine => 'http://superengine.com/ping?url=%s')
-    def ping_search_engines(*args)
-      require 'cgi/session'
+    def ping_search_engines(*args) # rubocop:disable Metrics/AbcSize,Metrics/MethodLength
       require 'open-uri'
       require 'timeout'
+      require 'uri'
 
       engines = args.last.is_a?(Hash) ? args.pop : {}
       unescaped_url = args.shift || sitemap_index_url
-      index_url = CGI.escape(unescaped_url)
+      index_url = URI.encode_www_form_component(unescaped_url)
 
       output("\n")
       output("Pinging with URL '#{unescaped_url}':")
       search_engines.merge(engines).each do |engine, link|
-        link = link % index_url
+        link %= index_url
         name = Utilities.titleize(engine.to_s)
         begin
-          Timeout::timeout(10) {
+          Timeout.timeout(10) do
             if URI.respond_to?(:open) # Available since Ruby 2.5
               URI.open(link)
             else
               open(link) # using Kernel#open became deprecated since Ruby 2.7. See https://bugs.ruby-lang.org/issues/15893
             end
-          }
+          end
           output("  Successful ping of #{name}")
         rescue Timeout::Error, StandardError => e
           output("Ping failed for #{name}: #{e.inspect} (URL #{link})")
@@ -384,26 +384,26 @@ module SitemapGenerator
     #
     # If both `filename` and `namer` are passed, set filename first so it
     # doesn't override the latter.
-    def set_options(opts={})
+    def set_options(opts = {})
       opts = opts.dup
-      %w(filename namer).each do |key|
-        if value = opts.delete(key.to_sym)
-          send("#{key}=", value)
+      %w[filename namer].each do |key|
+        if (value = opts.delete(key.to_sym))
+          send(:"#{key}=", value)
         end
       end
       opts.each_pair do |key, value|
-        send("#{key}=", value)
+        send(:"#{key}=", value)
       end
     end
 
     # Given +opts+, modify it and return it prepped for creating a new group from this LinkSet.
     # If <tt>:public_path</tt> is present in +opts+ it is removed because groups cannot
     # change the public path.
-    def options_for_group(opts)
+    def options_for_group(opts) # rubocop:disable Metrics/MethodLength
       opts = SitemapGenerator::Utilities.reverse_merge(opts,
-        :include_index => false,
-        :include_root => false,
-        :sitemap_index => sitemap_index
+        include_index: false,
+        include_root: false,
+        sitemap_index: sitemap_index
       )
       opts.delete(:public_path)
 
@@ -437,7 +437,7 @@ module SitemapGenerator
     # in an instance variable.
     def add_default_links
       @added_default_links = true
-      link_options = { :lastmod => Time.now, :priority => 1.0 }
+      link_options = { lastmod: Time.now, priority: 1.0 }
       if include_root?
         add('/', link_options)
       end
@@ -457,7 +457,8 @@ module SitemapGenerator
     # block passed to create() is empty the default links are still included in the
     # sitemap.
     def finalize_sitemap!
-      return if sitemap.finalized? || sitemap.empty? && @created_group
+      return if sitemap.finalized? || (sitemap.empty? && @created_group)
+
       add_default_links if !@added_default_links && !@created_group
       # This will finalize it.  We add to the index even if not creating an index because
       # the index keeps track of how many links are in our sitemaps and we need this info
@@ -470,6 +471,7 @@ module SitemapGenerator
     # been finalized.
     def finalize_sitemap_index!
       return if @protect_index || sitemap_index.finalized?
+
       sitemap_index.finalize!
       sitemap_index.write
     end
@@ -477,7 +479,7 @@ module SitemapGenerator
     # Return the interpreter linked to this instance.
     def interpreter
       require 'sitemap_generator/interpreter'
-      @interpreter ||= SitemapGenerator::Interpreter.new(:link_set => self)
+      @interpreter ||= SitemapGenerator::Interpreter.new(link_set: self)
     end
 
     # Reset this instance.  Keep the same options, but return to the same state
@@ -493,6 +495,7 @@ module SitemapGenerator
     # evaluated and some info output to STDOUT in a lazy fasion.
     def output(string)
       return unless verbose
+
       puts string
     end
 
@@ -576,28 +579,28 @@ module SitemapGenerator
       # Return a new +SitemapLocation+ instance with the current options included
       def sitemap_location
         SitemapGenerator::SitemapLocation.new(
-          :host => sitemaps_host,
-          :namer => namer,
-          :public_path => public_path,
-          :sitemaps_path => @sitemaps_path,
-          :adapter => @adapter,
-          :verbose => verbose,
-          :compress => @compress,
-          :max_sitemap_links => max_sitemap_links
+          host: sitemaps_host,
+          namer: namer,
+          public_path: public_path,
+          sitemaps_path: @sitemaps_path,
+          adapter: @adapter,
+          verbose: verbose,
+          compress: @compress,
+          max_sitemap_links: max_sitemap_links
         )
       end
 
       # Return a new +SitemapIndexLocation+ instance with the current options included
       def sitemap_index_location
         SitemapGenerator::SitemapLocation.new(
-          :host => sitemaps_host,
-          :namer => namer,
-          :public_path => public_path,
-          :sitemaps_path => @sitemaps_path,
-          :adapter => @adapter,
-          :verbose => verbose,
-          :create_index => @create_index,
-          :compress => @compress
+          host: sitemaps_host,
+          namer: namer,
+          public_path: public_path,
+          sitemaps_path: @sitemaps_path,
+          adapter: @adapter,
+          verbose: verbose,
+          create_index: @create_index,
+          compress: @compress
         )
       end
 
@@ -609,7 +612,7 @@ module SitemapGenerator
       # are in your sitemap.  If `false` an index file is never created.
       # If `:auto` an index file is created only if your sitemap has more than
       # one sitemap file.
-      def create_index=(value, force=false)
+      def create_index=(value, force = false)
         @create_index = value
         # Allow overriding the protected status of the index when we are creating a group.
         # Because sometimes we need to force an index in that case.  But generally we don't
@@ -658,8 +661,8 @@ module SitemapGenerator
 
       # Update the given attribute on the current sitemap index and sitemap file location objects.
       # But don't create the index or sitemap files yet if they are not already created.
-      def update_location_info(attribute, value, opts={})
-        opts = SitemapGenerator::Utilities.reverse_merge(opts, :include_index => !@protect_index)
+      def update_location_info(attribute, value, opts = {})
+        opts = SitemapGenerator::Utilities.reverse_merge(opts, include_index: !@protect_index)
         @sitemap_index.location[attribute] = value if opts[:include_index] && @sitemap_index && !@sitemap_index.finalized?
         @sitemap.location[attribute] = value if @sitemap && !@sitemap.finalized?
       end

data/lib/sitemap_generator/railtie.rb

@@ -1,7 +1,61 @@
+# frozen_string_literal: true
+
 module SitemapGenerator
   class Railtie < Rails::Railtie
+    # Top level options object to namespace all settings
+    config.sitemap = ActiveSupport::OrderedOptions.new
+
     rake_tasks do
-      load "tasks/sitemap_generator_tasks.rake"
+      load 'tasks/sitemap_generator_tasks.rake'
+    end
+
+    # Recognize existing Rails options as defaults for config.sitemap.*
+    initializer 'sitemap_generator.set_configs' do |app|
+      # routes.default_url_options takes precedence, falling back to configs
+      url_opts = (app.default_url_options || {})
+                 .with_defaults(config.try(:action_controller).try(:default_url_options) || {})
+                 .with_defaults(config.try(:action_mailer).try(:default_url_options) || {})
+                 .with_defaults(config.try(:active_job).try(:default_url_options) || {})
+
+      config.sitemap.default_host ||= ActionDispatch::Http::URL.full_url_for(url_opts) if url_opts.key?(:host)
+
+      # Rails defaults action_controller.asset_host and action_mailer.asset_host
+      # to the top-level config.asset_host so we get that for free here.
+      config.sitemap.sitemaps_host ||= [
+        config.try(:action_controller).try(:asset_host),
+        config.try(:action_mailer).try(:asset_host)
+      ].grep(String).first
+
+      config.sitemap.compress = config.try(:assets).try(:gzip) if config.sitemap.compress.nil?
+
+      config.sitemap.public_path ||= app.paths['public'].first
+
+      # "Compile" config.sitemap options onto the Sitemap class.
+      config.after_initialize do
+        ActiveSupport.on_load(:sitemap_generator, yield: true) do |sitemap|
+          config.sitemap.except(:adapter).each { |k, v| sitemap.public_send("#{k}=", v) }
+        end
+      end
+    end
+
+    # Allow setting the CONFIG_FILE without relying on env var;
+    # (e.g in config/application or environments/*.rb)
+    initializer 'sitemap_generator.config_file' do
+      if (config_file = config.sitemap.delete(:config_file).presence) && ENV['CONFIG_FILE'].blank?
+        ENV['CONFIG_FILE'] = config_file
+      end
+    end
+
+    # Allow lazily setting the adapter class without forcing an autoload.
+    # (ie. string or symbol name; or Callable (proc/lambda/etc))
+    initializer 'sitemap_generator.adapter' do |app|
+      config.to_prepare do
+        ActiveSupport.on_load(:sitemap_generator) do
+          self.adapter = Utilities.find_adapter app.config.sitemap.adapter
+        end
+      end
     end
   end
+
+  ActiveSupport.run_load_hooks(:sitemap_generator, Sitemap)
 end

data/lib/sitemap_generator/simple_namer.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module SitemapGenerator
   # A class for generating sitemap filenames.
   #
@@ -26,11 +28,11 @@ module SitemapGenerator
   #                is used, and subsequent names would be 'sitemap1.xml.gz', 'sitemap2.xml.gz', etc.
   #                In these examples the `base` string is assumed to be 'sitemap'.
   class SimpleNamer
-    def initialize(base, options={})
+    def initialize(base, options = {})
       @options = SitemapGenerator::Utilities.reverse_merge(options,
-        :zero => nil,  # identifies the marker for the start of the series
-        :extension => '.xml.gz',
-        :start => 1
+        zero: nil, # identifies the marker for the start of the series
+        extension: '.xml.gz',
+        start: 1
       )
       @base = base
       reset
@@ -63,7 +65,8 @@ module SitemapGenerator
 
     # Return this instance set to the previous name
     def previous
-      raise NameError, "Already at the start of the series" if start?
+      raise NameError, 'Already at the start of the series' if start?
+
       if @count <= @options[:start]
         @count = @options[:zero]
       else

data/lib/sitemap_generator/sitemap_location.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'sitemap_generator/helpers/number_helper'
 
 module SitemapGenerator
@@ -12,6 +14,7 @@ module SitemapGenerator
     [:host, :adapter].each do |method|
       define_method(method) do
         raise SitemapGenerator::SitemapError, "No value set for #{method}" unless self[method]
+
         self[method]
       end
     end
@@ -45,7 +48,7 @@ module SitemapGenerator
     #   stripped from the filename.  If `:all_but_first`, only the `.gz` extension of the first
     #   filename is stripped off.  If `true` the extensions are left unchanged.
     # * <tt>max_sitemap_links</tt> - The maximum number of links to put in each sitemap.
-    def initialize(opts={})
+    def initialize(opts = {})
       SitemapGenerator::Utilities.assert_valid_keys(opts, [
         :adapter,
         :public_path,
@@ -66,14 +69,14 @@ module SitemapGenerator
       # sitemap index files.  However, this greatly eases testing, so I'm leaving it in
       # for now.
       if !opts[:filename] && !opts[:namer]
-        opts[:namer] = SitemapGenerator::SimpleNamer.new(:sitemap, :start => 2, :zero => 1)
+        opts[:namer] = SitemapGenerator::SimpleNamer.new(:sitemap, start: 2, zero: 1)
       end
       opts[:verbose] = !!opts[:verbose]
       self.merge!(opts)
     end
 
     # Return a new Location instance with the given options merged in
-    def with(opts={})
+    def with(opts = {})
       self.merge(opts)
     end
 
@@ -106,9 +109,10 @@ module SitemapGenerator
     # If using a namer once the filename has been retrieved from the namer its
     # value is locked so that it is unaffected by further changes to the namer.
     def filename
-      raise SitemapGenerator::SitemapError, "No filename or namer set" unless self[:filename] || self[:namer]
+      raise SitemapGenerator::SitemapError, 'No filename or namer set' unless self[:filename] || self[:namer]
+
       unless self[:filename]
-        self.send(:[]=, :filename, self[:namer].to_s, :super => true)
+        self.send(:[]=, :filename, +self[:namer].to_s, super: true)
 
         # Post-process the filename for our compression settings.
         # Strip the `.gz` from the extension if we aren't compressing this file.
@@ -148,7 +152,7 @@ module SitemapGenerator
     end
 
     # If you set the filename, clear the namer and vice versa.
-    def []=(key, value, opts={})
+    def []=(key, value, opts = {})
       if !opts[:super]
         case key
         when :namer
@@ -172,12 +176,12 @@ module SitemapGenerator
       filesize = number_to_human_size(self.filesize)
       width = self.class::PATH_OUTPUT_WIDTH
       path = SitemapGenerator::Utilities.ellipsis(self.path_in_public, width)
-      "+ #{('%-'+width.to_s+'s') % path} #{'%10s' % link_count} links / #{'%10s' % filesize}"
+      "+ #{('%-' + width.to_s + 's') % path} #{'%10s' % link_count} links / #{'%10s' % filesize}"
     end
   end
 
   class SitemapIndexLocation < SitemapLocation
-    def initialize(opts={})
+    def initialize(opts = {})
       if !opts[:filename] && !opts[:namer]
         opts[:namer] = SitemapGenerator::SimpleNamer.new(:sitemap)
       end
@@ -198,7 +202,7 @@ module SitemapGenerator
       filesize = number_to_human_size(self.filesize)
       width = self.class::PATH_OUTPUT_WIDTH - 3
       path = SitemapGenerator::Utilities.ellipsis(self.path_in_public, width)
-      "+ #{('%-'+width.to_s+'s') % path} #{'%10s' % link_count} sitemaps / #{'%10s' % filesize}"
+      "+ #{('%-' + width.to_s + 's') % path} #{'%10s' % link_count} sitemaps / #{'%10s' % filesize}"
     end
   end
 end

data/lib/sitemap_generator/tasks.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 # require this file to load the tasks
 require 'rake'
 
@@ -8,46 +10,32 @@ namespace :sitemap do
   # Require sitemap_generator only.  When installed as a plugin the require will fail, so in
   # that case, load the environment first.
   task :require do
-    begin
-      require 'sitemap_generator'
-    rescue LoadError => e
-      if defined?(Rails::VERSION)
-        Rake::Task['sitemap:require_environment'].invoke
-      else
-        raise e
-      end
-    end
-  end
-
-  # Require sitemap_generator after loading the Rails environment.  We still need the require
-  # in case we are installed as a gem and are setup to not automatically be required.
-  task :require_environment do
-    if defined?(Rails::VERSION)
-      Rake::Task['environment'].invoke
-    end
     require 'sitemap_generator'
   end
+  # In a Rails app, we need to boot Rails.
+  # Ensure gem is required in case it wasn't automatically loaded.
+  Rake::Task[:require].enhance([:environment]) if defined?(Rails)
 
-  desc "Install a default config/sitemap.rb file"
-  task :install => ['sitemap:require'] do
+  desc 'Install a default config/sitemap.rb file'
+  task install: :require do
     SitemapGenerator::Utilities.install_sitemap_rb(verbose)
   end
 
-  desc "Delete all Sitemap files in public/ directory"
-  task :clean => ['sitemap:require'] do
+  desc 'Delete all Sitemap files in public/ directory'
+  task clean: :require do
     SitemapGenerator::Utilities.clean_files
   end
 
-  desc "Generate sitemaps and ping search engines."
-  task :refresh => ['sitemap:create'] do
+  desc 'Generate sitemaps and ping search engines.'
+  task refresh: :create do
     SitemapGenerator::Sitemap.ping_search_engines
   end
 
   desc "Generate sitemaps but don't ping search engines."
-  task 'refresh:no_ping' => ['sitemap:create']
+  task 'refresh:no_ping' => :create
 
   desc "Generate sitemaps but don't ping search engines.  Alias for refresh:no_ping."
-  task :create => ['sitemap:require_environment'] do
-    SitemapGenerator::Interpreter.run(:config_file => ENV["CONFIG_FILE"], :verbose => verbose)
+  task create: :require do
+    SitemapGenerator::Interpreter.run(config_file: ENV['CONFIG_FILE'], verbose: verbose)
   end
 end

data/lib/sitemap_generator/templates.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module SitemapGenerator
   # Provide convenient access to template files.  E.g.
   #
@@ -7,17 +9,18 @@ module SitemapGenerator
   # Define an accessor method for each template file.
   class Templates
     FILES = {
-      :sitemap_sample =>  'sitemap.rb',
+      sitemap_sample: 'sitemap.rb',
     }
 
     # Dynamically define accessors for each key defined in <tt>FILES</tt>
-    attr_accessor(*FILES.keys)
-    FILES.keys.each do |name|
-      eval <<-END
+    attr_writer(*FILES.keys)
+
+    FILES.each_key do |name|
+      eval(<<-ACCESSOR, binding, __FILE__, __LINE__ + 1)
         define_method(:#{name}) do
           @#{name} ||= read_template(:#{name})
         end
-      END
+      ACCESSOR
     end
 
     def initialize(root = SitemapGenerator.root)