simple_whatweb 0.2.0 → 0.2.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 74949bba132a54310d0734e532ea57ea10df193d42748826bb1ec8f0a94ce704
-  data.tar.gz: dff1cb823b0c7f36e38524eeba7433267d06120f23bd7c64df696d917b93a6a6
+  metadata.gz: fa93fc00fb8c3719932816eda5f66233849319c2c97536b74a9fd3590864f3ab
+  data.tar.gz: a24e458f6488eeb31b5ca814b14af6662363b807d8f72f57053fe364f8f6aa41
 SHA512:
-  metadata.gz: 78c15492948ebae1399a703f0e57072881e0fbf57451f4f8ac79ef350ca2f325108618207724f0eee36db8479a3c91b3effaeab7b4e66b051d7f8e54aad14cf3
-  data.tar.gz: a3e404817a1d41a34b54972bc534a4ffc86cdb6b4112c6ab23468b1f4410da5fa9e1db26713f05ddcdc64ac5540422530cfb8a4b9bd0a9d5441101fd4e7c4669
+  metadata.gz: d42ceb9c118d0081712d4e5551b6899d0ec774af54ed3cf623744dbb59be8321ecdd785b32f3531a0a17ba7e4d0f77daa976bbf3fd1ea958b9bc84750a72d0a9
+  data.tar.gz: 98609d498e9a948cf813d80031864f8d346905fecfa3e551497e970b2dff8f7525bb49b24d135baa67b97c3af523fc2b7701d00ab7fd3d7d1c95d721c8df9824

data/SimpleWhatWeb.gemspec

@@ -32,7 +32,7 @@ Gem::Specification.new do |spec|
   spec.add_development_dependency "webmock", "~> 3.4"
 
   spec.add_dependency "http", "~> 3.3"
+  spec.add_dependency "oga", "~> 2.15"
   spec.add_dependency "require_all", "~> 2.0"
-  spec.add_dependency "sanitize", "~> 4.6"
   spec.add_dependency "thor", "~> 0.19"
 end

data/lib/whatweb/helper.rb

@@ -1,7 +1,7 @@
 # frozen_string_literal: true
 
 require "digest/md5"
-require "sanitize"
+require "oga"
 
 module WhatWeb
   module Helper
@@ -10,8 +10,11 @@ module WhatWeb
         Digest::MD5.hexdigest(body.to_s)
       end
 
-      def sanitized_body
-        Sanitize.document(body.to_s, elements: ["html"])
+      def text
+        doc = Oga.parse_html(body.to_s.force_encoding('UTF-8'))
+        path = /\A<body(?:\s|>)/i.match?(body.to_s) ? '/html/body' : '/html/body/node()'
+        nodes = doc.xpath(path)
+        nodes.map(&:text).join
       end
 
       def tag_pattern

data/lib/whatweb/matcher/ghdb.rb

@@ -45,10 +45,10 @@ module WhatWeb
           # does it start with a - ?
           if w[0] == '-'
             # reverse true/false if it begins with a -
-            !target.sanitized_body.match? /#{Regexp.escape(w[1..-1])}/i
+            !target.text.match? /#{Regexp.escape(w[1..-1])}/i
           else
             w = w[1..-1] if w[0] == '+' # if it starts with +, ignore the 1st char
-            target.sanitized_body.match? /#{Regexp.escape(w)}/i
+            target.text.match? /#{Regexp.escape(w)}/i
           end
         end
       end

data/lib/whatweb/plugins/anygate.rb

@@ -1,4 +1,3 @@
-# coding: ascii-8bit
 # frozen_string_literal: true
 
 ##

data/lib/whatweb/plugins/bm-classifieds.rb

@@ -1,4 +1,3 @@
-# coding: ascii-8bit
 # frozen_string_literal: true
 
 ##

data/lib/whatweb/plugins/fidion-cms.rb

@@ -1,4 +1,3 @@
-# coding: ascii-8bit
 # frozen_string_literal: true
 
 ##

data/lib/whatweb/plugins/mihalism-multi-host.rb

@@ -1,4 +1,3 @@
-# coding: ascii-8bit
 # frozen_string_literal: true
 
 ##

data/lib/whatweb/plugins/netsnap-web-camera.rb

@@ -1,4 +1,3 @@
-# coding: ascii-8bit
 # frozen_string_literal: true
 
 ##

data/lib/whatweb/plugins/rvi-camera.rb

@@ -1,4 +1,3 @@
-# coding: ascii-8bit
 # frozen_string_literal: true
 
 ##

data/lib/whatweb/plugins/s-cms.rb

@@ -1,4 +1,3 @@
-# coding: ascii-8bit
 # frozen_string_literal: true
 
 ##

data/lib/whatweb/plugins/samphpweb.rb

@@ -1,4 +1,3 @@
-# coding: ascii-8bit
 # frozen_string_literal: true
 
 ##

data/lib/whatweb/plugins/snografx.rb

@@ -1,4 +1,3 @@
-# coding: ascii-8bit
 # frozen_string_literal: true
 
 ##

data/lib/whatweb/plugins/title.rb

@@ -12,7 +12,7 @@
 # Version 0.2
 # removed :certainty=>100
 
-require "nokogiri"
+require "oga"
 
 WhatWeb::Plugin.define "Title" do
   @author = "Andrew Horton"
@@ -22,8 +22,8 @@ WhatWeb::Plugin.define "Title" do
   def passive(target)
     m = []
 
-    html = Nokogiri.parse(target.body)
-    title = html.css("title")
+    html = Oga.parse_html(target.body)
+    title = html.at_css("title")
     if title
       # Give warining if title element contains newline(s)
       m << { name: "WARNING", module: "Title element contains newline(s)!" } if title.text.include? "\n"

data/lib/whatweb/target.rb

@@ -40,8 +40,8 @@ module WhatWeb
       @tag_pattern ||= response.tag_pattern
     end
 
-    def sanitized_body
-      @sanitized_body ||= response.sanitized_body
+    def text
+      @text ||= response.text
     end
 
     def self.meta_refresh_regex

data/lib/whatweb/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 module WhatWeb
-  VERSION = "0.2.0".freeze
+  VERSION = "0.2.1"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: simple_whatweb
 version: !ruby/object:Gem::Version
-  version: 0.2.0
+  version: 0.2.1
 platform: ruby
 authors:
 - Manabu Niseki
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2018-08-19 00:00:00.000000000 Z
+date: 2018-09-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -109,33 +109,33 @@ dependencies:
       - !ruby/object:Gem::Version
         version: '3.3'
 - !ruby/object:Gem::Dependency
-  name: require_all
+  name: oga
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '2.0'
+        version: '2.15'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '2.0'
+        version: '2.15'
 - !ruby/object:Gem::Dependency
-  name: sanitize
+  name: require_all
   requirement: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.6'
+        version: '2.0'
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
     requirements:
     - - "~>"
       - !ruby/object:Gem::Version
-        version: '4.6'
+        version: '2.0'
 - !ruby/object:Gem::Dependency
   name: thor
   requirement: !ruby/object:Gem::Requirement