legitbot 1.0.1 → 1.0.2

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 1596ed8c3809fc3c3068f14d8bbbcf84232286fd7157fe724ba9515d195259cd
-  data.tar.gz: f0078d0404d752550adeeaa9c64954cbcec57fef3d0b37c98bc50d765c29bc1f
+  metadata.gz: c0b43dcf4386a8427415729bf9ba5edfe3c0efc24a002bb4301ff98d43c4ee0d
+  data.tar.gz: 9f18f03e613473a97cda23e8663d44da10cfd32b3fd271e1ffd5ee20cfdef48d
 SHA512:
-  metadata.gz: 5640ae1e351bbd697325cd0bed0bdac45c3726e7b31923b7b5c8f51859ed025d4eb368d937a2b3bf9ef6c150d83418f8e019304dec3bc1b010572a4e1598c661
-  data.tar.gz: bf326da52d3adf1b2cfa3693b51cc763daef54c9b175cfb91b437b213722a0f578565db553f1240b6d6ee47d159bb418fdcc1db1f95cb14696b2bfb52e50b75d
+  metadata.gz: 73ad2e000c4513880bc5558c7fe5f7c7c04d01b1a820ef7b12541c8c9d6cc5dddd20c785d67ef489d5b00664632c6b88a805c3bb1e2a060517126c7d7292d661
+  data.tar.gz: a578a30a512495523e906ec1d6a650f4f472a58856caa542714d3e57a0b99f764ee015f11624c4a20cae79a3975eb41eaf8ced106977dd994981d71c288b5653

data/.github/workflows/build.yml

@@ -1,6 +1,6 @@
 name: build
 
-on: [push]
+on: [pull_request, push]
 
 jobs:
   test:

data/.rubocop.yml

@@ -1,2 +1,3 @@
 AllCops:
   CacheRootDirectory: 'vendor'
+  NewCops: enable

data/README.md

@@ -44,6 +44,7 @@ end
 ## Supported
 
 * [Ahrefs](https://ahrefs.com/robot)
+* [Alexa](https://support.alexa.com/hc/en-us/articles/360046707834-What-are-the-IP-addresses-for-Alexa-s-Certify-and-Site-Audit-crawlers-)
 * [Applebot](https://support.apple.com/en-us/HT204683)
 * [Baidu spider](http://help.baidu.com/question?prod_en=master&class=498&id=1000973)
 * [Bingbot](https://blogs.bing.com/webmaster/2012/08/31/how-to-verify-that-bingbot-is-bingbot/)

data/legitbot.gemspec

@@ -20,7 +20,7 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency 'bump', '~> 0.8', '>= 0.8.0'
   spec.add_development_dependency 'minitest', '~> 5.1', '>= 5.1.0'
   spec.add_development_dependency 'rake', '~> 12.3', '>= 12.3.0'
-  spec.add_development_dependency 'rubocop', '~> 0.74', '>= 0.74.0'
+  spec.add_development_dependency 'rubocop', '~> 0.90', '< 0.91'
 
   spec.files = `git ls-files`.split($INPUT_RECORD_SEPARATOR)
   spec.rdoc_options = ['--charset=UTF-8']

data/lib/legitbot.rb

@@ -4,6 +4,7 @@ require_relative 'legitbot/legitbot'
 require_relative 'legitbot/botmatch'
 
 require_relative 'legitbot/ahrefs'
+require_relative 'legitbot/alexa'
 require_relative 'legitbot/apple'
 require_relative 'legitbot/baidu'
 require_relative 'legitbot/bing'

data/lib/legitbot/alexa.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+module Legitbot # :nodoc:
+  # https://support.alexa.com/hc/en-us/articles/360046707834-What-are-the-IP-addresses-for-Alexa-s-Certify-and-Site-Audit-crawlers-
+  # https://support.alexa.com/hc/en-us/articles/200462340
+  # https://support.alexa.com/hc/en-us/articles/200450194
+  class Alexa < BotMatch
+    ip_ranges %w[
+      52.86.176.3
+      52.4.48.181
+      52.2.182.169
+      52.86.185.29
+    ]
+  end
+
+  rule Legitbot::Alexa, %w[Alexabot ia_archiver]
+end

data/lib/legitbot/legitbot.rb

@@ -18,7 +18,7 @@ module Legitbot
   # otherwise.
   # :yields: a found bot
   #
-  # rubocop:disable Metrics/CyclomaticComplexity
+  # rubocop:disable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/PerceivedComplexity
   def self.bot(user_agent, ip)
     bots = @rules
            .select { |rule| rule[:fragments].any? { |f| user_agent.index f } }
@@ -33,7 +33,7 @@ module Legitbot
       selected
     end
   end
-  # rubocop:enable Metrics/CyclomaticComplexity
+  # rubocop:enable Metrics/AbcSize, Metrics/CyclomaticComplexity, Metrics/PerceivedComplexity
 
   def self.rule(clazz, fragments)
     @rules << { class: clazz, fragments: fragments }

data/lib/legitbot/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module Legitbot
-  VERSION = '1.0.1'
+  VERSION = '1.0.2'
 end

data/test/alexa_test.rb

@@ -0,0 +1,36 @@
+# frozen_string_literal: true
+
+require 'minitest/autorun'
+require 'legitbot'
+
+class AlexaTest < Minitest::Test
+  def test_malicious_ip
+    ip = '149.210.164.47'
+    match = Legitbot::Alexa.new ip
+    assert !match.valid?, msg: "#{ip} is not a real Alexa IP"
+  end
+
+  def test_valid_ip
+    ip = '52.86.176.3'
+    match = Legitbot::Alexa.new ip
+    assert match.valid?, msg: "#{ip} is a valid Alexa IP"
+  end
+
+  def test_malicious_ua
+    bot = Legitbot.bot(
+      'Mozilla/5.0 (compatible; Alexabot/1.0; +http://www.alexa.com/help/certifyscan; certifyscan@alexa.com)',
+      '149.210.164.47'
+    )
+    assert bot, msg: 'Alexa detected from User-Agent'
+    assert !bot.valid?, msg: 'Not a valid Alexa'
+  end
+
+  def test_valid_ua
+    bot = Legitbot.bot(
+      'Mozilla/5.0 (compatible; Alexabot/1.0; +http://www.alexa.com/help/certifyscan; certifyscan@alexa.com)',
+      '52.86.176.3'
+    )
+    assert bot, msg: 'Alexa detected from User-Agent'
+    assert bot.valid?, msg: 'Valid Alexa'
+  end
+end

data/test/facebook_test.rb

@@ -33,7 +33,7 @@ class FacebookTest < Minitest::Test
     assert match.fake?, msg: "#{ip} is a fake Facebook IP"
   end
 
-  # rubocop:disable Metrics/AbcSize, Layout/LineLength, Metrics/MethodLength
+  # rubocop:disable Layout/LineLength, Metrics/MethodLength
   def test_user_agent
     Legitbot.bot(
       'facebookexternalhit/1.1 (+http://www.facebook.com/externalhit_uatext.php)',
@@ -59,5 +59,5 @@ class FacebookTest < Minitest::Test
       assert bot.fake?, msg: 'fake Facebook'
     end
   end
-  # rubocop:enable Metrics/AbcSize, Layout/LineLength, Metrics/MethodLength
+  # rubocop:enable Layout/LineLength, Metrics/MethodLength
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: legitbot
 version: !ruby/object:Gem::Version
-  version: 1.0.1
+  version: 1.0.2
 platform: ruby
 authors:
 - Alexander Azarov
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2020-07-28 00:00:00.000000000 Z
+date: 2020-09-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: augmented_interval_tree
@@ -116,20 +116,20 @@ dependencies:
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.74'
-    - - ">="
+        version: '0.90'
+    - - "<"
       - !ruby/object:Gem::Version
-        version: 0.74.0
+        version: '0.91'
   type: :development
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '0.74'
-    - - ">="
+        version: '0.90'
+    - - "<"
       - !ruby/object:Gem::Version
-        version: 0.74.0
+        version: '0.91'
 description: Does Web request come from a real search engine or from an impersonating
   agent?
 email: self@alaz.me
@@ -148,6 +148,7 @@ files:
 - legitbot.gemspec
 - lib/legitbot.rb
 - lib/legitbot/ahrefs.rb
+- lib/legitbot/alexa.rb
 - lib/legitbot/apple.rb
 - lib/legitbot/baidu.rb
 - lib/legitbot/bing.rb
@@ -165,6 +166,7 @@ files:
 - lib/legitbot/version.rb
 - lib/legitbot/yandex.rb
 - test/ahrefs_test.rb
+- test/alexa_test.rb
 - test/apple_as_google_test.rb
 - test/apple_test.rb
 - test/botmatch_test.rb
@@ -205,6 +207,7 @@ test_files:
 - test/legitbot/validators/domains_test.rb
 - test/legitbot/validators/ip_ranges_test.rb
 - test/pinterest_test.rb
+- test/alexa_test.rb
 - test/ahrefs_test.rb
 - test/apple_test.rb
 - test/apple_as_google_test.rb