From 21fd4d787848fa082ce1e21845d4728933b99fd0 Mon Sep 17 00:00:00 2001
From: MothOnMars <mothonmars@gmail.com>
Date: Wed, 24 Jan 2018 08:57:32 -0800
Subject: [PATCH] add #sitemaps method

---
 README.rdoc          |  1 +
 lib/robotex.rb       | 13 ++++++++-
 spec/robotex_spec.rb | 65 +++++++++++++++++++++++++++++++-------------
 3 files changed, 59 insertions(+), 20 deletions(-)

diff --git a/README.rdoc b/README.rdoc
index 83c96aa..5326bdb 100644
--- a/README.rdoc
+++ b/README.rdoc
@@ -8,6 +8,7 @@ Usage:
     robotex = Robotex.new "My User Agent"
     robotex.allowed?("http://www.example.com/foo")
     robotex.delay!("http://www.example.com/foo") # wait until any specified Crawl-Delay has passed
+    robotex.sitemaps("http://www.example.com/") # return an array of sitemap urls
 
 == Acknowledgements
 
diff --git a/lib/robotex.rb b/lib/robotex.rb
index bf186ab..0d78364 100644
--- a/lib/robotex.rb
+++ b/lib/robotex.rb
@@ -15,6 +15,8 @@ class Robotex
   
   class ParsedRobots
     
+    attr_reader :sitemaps
+
     def initialize(uri, user_agent)
       io = Robotex.get_robots_txt(uri, user_agent)
       
@@ -25,6 +27,7 @@ def initialize(uri, user_agent)
       @disallows = {}
       @allows = {}
       @delays = {}
+      @sitemaps = []
       agent = /.*/
       io.each do |line|
         next if line =~ /^\s*(#.*|$)/
@@ -43,6 +46,8 @@ def initialize(uri, user_agent)
             @disallows[agent] << to_regex(value)
           when "crawl-delay"
             @delays[agent] = value.to_i
+          when "sitemap"
+            @sitemaps << URI.join(uri, value).to_s
         end
       end
       
@@ -148,5 +153,11 @@ def delay!(uri)
     sleep delay - (Time.now - @last_accessed) if !!delay
     @last_accessed = Time.now
   end
-  
+
+  #
+  # Returns an array of the sitemap urls specified in robots.txt
+  #
+  def sitemaps(uri)
+    parse_host(uri).sitemaps
+  end
 end
diff --git a/spec/robotex_spec.rb b/spec/robotex_spec.rb
index d8b4388..51963ca 100644
--- a/spec/robotex_spec.rb
+++ b/spec/robotex_spec.rb
@@ -1,25 +1,30 @@
 require 'spec_helper'
 
 describe Robotex do
+  let(:robots) do
+    <<~ROBOTS
+      User-Agent: msnbot
+      Crawl-Delay: 20
 
-  before(:all) do
+      User-Agent: bender
+      Disallow: /my_shiny_metal_ass
+
+      User-Agent: *
+      Disallow: /login
+      Allow: /
+
+      Disallow: /locked
+      Allow: /locked
+    ROBOTS
+  end
+
+  let(:response) do
+    { body: robots, content_type: 'text/plain', status: [200, "OK"] }
+  end
+
+  before do
     FakeWeb.allow_net_connect = false
-    robots = <<-END
-User-Agent: msnbot
-Crawl-Delay: 20
-
-User-Agent: bender
-Disallow: /my_shiny_metal_ass
-
-User-Agent: *
-Disallow: /login
-Allow: /
-
-Disallow: /locked
-Allow: /locked
-END
-    options = {:body => robots, :content_type => 'text/plain', :status => [200, "OK"]}
-    FakeWeb.register_uri(:get, SPEC_DOMAIN + 'robots.txt', options)
+    FakeWeb.register_uri(:get, SPEC_DOMAIN + 'robots.txt', response)
   end
 
   describe '#initialize' do
@@ -73,6 +78,7 @@
         robotex = Robotex.new
         robotex.delay(SPEC_DOMAIN).should be_nil 
       end
+    end
 
     context 'when Crawl-Delay is specified for the user-agent' do
       it 'returns the delay as a Fixnum' do
@@ -80,8 +86,29 @@
         robotex.delay(SPEC_DOMAIN).should == 20
       end
     end
-    end
   end
 
-end
+  describe '#sitemaps' do
+    let(:robots) do
+      <<~ROBOTS
+        Sitemap: http://www.example.com/sitemap_1.xml
+        Sitemap: http://www.example.com/sitemap_2.xml
+      ROBOTS
+    end
+
+    it 'returns an array of sitemaps' do
+      robotex = Robotex.new
+      robotex.sitemaps(SPEC_DOMAIN).should == %w[http://www.example.com/sitemap_1.xml
+                                                 http://www.example.com/sitemap_2.xml]
+    end
 
+    context 'when the sitemap url is relative' do
+      let(:robots) { 'Sitemap: /relative.xml' }
+
+      it 'returns the sitemap' do
+        robotex = Robotex.new
+        robotex.sitemaps(SPEC_DOMAIN).should == ['http://www.example.com/relative.xml']
+      end
+    end
+  end
+end