Update CHANGELOG.md

CHANGELOG.md

@@ -1,5 +1,20 @@
 ## [Unreleased]
 
+## [0.5.0] - 2024-12-24
+
+- Fix bug with Disallow rule containing empty line
+
+## [0.4.0] - 2024-10-31
+
+- Ensure VERISON is available
+
+## [0.3.0] - 2023-09-18
+
+- Only return unique sitemaps.
+
+## [0.2.0] - 2023-09-10
+
+- Correctly handle multiple sitemaps + tests.
 ## [0.1.0] - 2023-09-09
 
 - Initial release

lib/probot.rb

@@ -2,6 +2,7 @@
 
 require "uri"
 require "net/http"
+require_relative "probot/version"
 
 # https://moz.com/learn/seo/robotstxt
 # https://stackoverflow.com/questions/45293419/order-of-directives-in-robots-txt-do-they-overwrite-each-other-or-complement-ea
@@ -30,6 +31,7 @@ class Probot
     @current_agents = ["*"]
     @current_agents.each { |agent| @rules[agent] ||= {"disallow" => [], "allow" => [], "crawl_delay" => 0} }
     @sitemaps = []
+
     @site = URI(data) if data.start_with?("http")
     @doc = @site.nil? ? data : fetch_robots_txt(@site)
     parse(@doc)
@@ -91,7 +93,9 @@ class Probot
 
       # All Regex characters are escaped, then we unescape * and $ as they may used in robots.txt
       if data.allow? || data.disallow?
-        @current_agents.each { |agent| rules[agent][data.key] << Regexp.new(Regexp.escape(data.value).gsub('\*', ".*").gsub('\$', "$")) }
+        @current_agents.each do |agent|
+          rules[agent][data.key] << Regexp.new(Regexp.escape(data.value).gsub('\*', ".*").gsub('\$', "$")) unless data.value.nil?
+        end
 
         # When user-agent strings are found on consecutive lines, they are considered to be part of the same record. Google ignores crawl_delay.
         subsequent_agent = false
@@ -108,6 +112,7 @@ class Probot
         sitemap_uri = URI(data.value)
         sitemap_uri = sitemap_uri.host.nil? ? URI.join(*[site, sitemap_uri].compact) : sitemap_uri
         @sitemaps << sitemap_uri.to_s
+        @sitemaps.uniq!
         next
       end
 
@@ -126,6 +131,8 @@ class Probot
 
     def clean_value = raw_value.split("#").first&.strip
 
+    def clean_url = clean_value&.then { URI(_1).to_s }
+
     def agent? = key == "user-agent"
 
     def disallow? = key == "disallow"
@@ -138,11 +145,13 @@ class Probot
 
     def value
       return clean_value.to_f if crawl_delay?
-      return URI(clean_value).to_s if disallow? || allow?
+      return clean_url if disallow? || allow?
 
       raw_value
     rescue URI::InvalidURIError
       raw_value
+    rescue ArgumentError
+      raw_value
     end
   end
 

lib/probot/version.rb

@@ -1,3 +1,3 @@
 class Probot
-  VERSION = "0.2.0"
+  VERSION = "0.5.0"
 end

probot.gemspec

@@ -13,6 +13,7 @@ Gem::Specification.new do |spec|
   spec.homepage = "http://github.com/dkam/probot"
   spec.license = "MIT"
   spec.required_ruby_version = ">= 3.0"
+  spec.platform = Gem::Platform::RUBY
 
   spec.metadata["homepage_uri"] = spec.homepage
   spec.metadata["source_code_uri"] = "http://github.com/dkam/probot"
@@ -29,4 +30,5 @@ Gem::Specification.new do |spec|
   spec.bindir = "exe"
   spec.executables = spec.files.grep(%r{\Aexe/}) { |f| File.basename(f) }
   spec.require_paths = ["lib"]
+  spec.add_development_dependency "debug"
 end

test/test_helper.rb

@@ -2,6 +2,5 @@
 
 $LOAD_PATH.unshift File.expand_path("../lib", __dir__)
 require "probot"
-require "probot/version" # for testing the version number - otherwise the gemspec does it.
 
 require "minitest/autorun"

test/test_probot.rb

@@ -95,6 +95,19 @@ class TestProbot < Minitest::Test
           crawl_delay: 0
         }
       ]
+    },
+    {
+      txt: %("User-agent: *\nDisallow: /wp/wp-admin/\nAllow: /wp/wp-admin/admin-ajax.php\n\nUser-agent: *\nDisallow: /wp-content/uploads/wpo/wpo-plugins-tables-list.json\n\n# START YOAST BLOCK\n# ---------------------------\nUser-agent: *\nDisallow:\n\nSitemap: https://prhinternationalsales.com/sitemap_index.xml\n# ---------------------------\n# END YOAST BLOCK"),
+      sitemaps: ["https://prhinternationalsales.com/sitemap_index.xml"],
+      found_agents: ["*"],
+      tests: [
+        {
+          agent: "*",
+          allowed: ["/wp/wp-admin/admin-ajax.php"],
+          disallowed: ["/wp/wp-admin/", "/wp-content/uploads/wpo/wpo-plugins-tables-list.json"],
+          crawl_delay: 0
+        }
+      ]
     }
   ].freeze
 
@@ -131,7 +144,9 @@ class TestProbot < Minitest::Test
 
   def test_empty_allow_disallow
     assert Probot.new(%(User-agent: *\nAllow:)).rules.dig("*", "allow").empty?
+    assert Probot.new(%(User-agent: *\nAllow:\n\n)).rules.dig("*", "allow").empty?
     assert Probot.new(%(User-agent: *\nDisallow:)).rules.dig("*", "disallow").empty?
+    assert Probot.new(%(User-agent: *\nDisallow:\n\n)).rules.dig("*", "disallow").empty?
   end
 
   def test_consecutive_user_agents