validate-website 1.9.1 → 1.11.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 60082f1dc222de92a3465b0487e6047597d7605fd2199e15f246a5930ddd5ea3
-  data.tar.gz: 1fd7bd74bc0b58bbb16e82ae396309db5638362f6f9853d80161849299ff28d8
+  metadata.gz: 7208a9c728f7798aecc7a10bec8a3537940a6857e3eb4f6612efa551f2871a94
+  data.tar.gz: 25e5112b42c14a6ed86f0271fba1bb9ada63c1d92cc75414a227716289e241f8
 SHA512:
-  metadata.gz: 3ab5fa6da0d050fc39acabdd4cb26a764acfdb609df4786cb22f55af5d2f500e0638a55e67505cb1d6e1595e0c04ede5ad4ed88b46bf078bc09f58585b7f481c
-  data.tar.gz: '0199501f535de09d55bc9a8f28549f4c01a94d9444f12732ad02fc503e36158daa1535a61e9b349ea462eb4799bebb5a599417e93dbe2b0abd92b0e0319abb28'
+  metadata.gz: 95962fe71e144cc5c8f244e1960556e56c9956bdcd5852e194532ee16b40dfbda5ba39db75169651ddd4048afc79311fdee2aa1087dddc3388aaa894c4dc27fe
+  data.tar.gz: 6a6e8185f8a359e0482c0ef5893b28d307a31be560a2e83bfd19f88f042cfbd3ec37e244f602e5ea4a6876afce9b8d2a38f98f645452168b752c03ee62fdc70d

data/History.md

@@ -1,4 +1,48 @@
 
+1.11.1 / 2021-01-10
+===================
+
+  * Add nokogumbo doc
+  * nokogumbo support for ruby engine
+  * Add support for nokogumbo
+  * Add jruby to github actions
+  * Merge pull request #24 from @marocchino / patch-1
+    * Some minor improvements
+  * Remove travis ci
+  * Init github actions ci
+
+1.11.0 / 2021-01-08
+===================
+
+  * Merge pull request #23 from @marocchino / ruby-3-support
+    * Use webrick's escape instead of encode_www_form_component
+    * Support ruby 3
+  * Fix doc for ValidateWebsite::Core initialize
+  * Switch to gitlab ci and remove 2.{3,4} support
+  * Update rubocop to 0.76.0
+
+1.10.0 / 2020-07-03
+==================
+
+  * Fix build for Ruby 2.3 and 2.4
+  * Remove rbx-3 from build
+  * Remove minitest-focus and fix minitest 6 warnings
+  * Fix html5_validator option and change html5_validator_service_url
+  * Add Ruby 2.7 to CI and update jruby
+  * Update rubocop and fix offences
+  * Remove Ruby 2.2 support and update rubocop
+
+1.9.3 / 2019-04-11
+==================
+
+  * Update tidy_ffi to 1.0
+  * Avoid testing tidy bug with js
+
+1.9.2 / 2019-03-09
+==================
+
+  * Load schema when needed instead of boot
+
 1.9.1 / 2019-03-05
 ==================
 

data/LICENSE

@@ -1,6 +1,6 @@
 The MIT License
 
-Copyright (c) 2009-2019 Laurent Arnoud <laurent@spkdev.net>
+Copyright (c) 2009-2021 Laurent Arnoud <laurent@spkdev.net>
 
 Permission is hereby granted, free of charge, to any person obtaining
 a copy of this software and associated documentation files (the

data/README.md

@@ -95,6 +95,11 @@ If the libtidy5 is found on your system this will be the default to validate
 your html5 document. This does not depend on a tier service everything is done
 locally.
 
+#### Nokogumbo
+
+Nokogumbo can validate html5 document without tier service but reports less
+errors than tidy.
+
 #### Validator.nu web service
 
 When `--html5-validator nu` option is used HTML5 support is done by using the
@@ -145,12 +150,12 @@ See [GitHub](https://github.com/spk/validate-website/graphs/contributors).
 
 The MIT License
 
-Copyright (c) 2009-2019 Laurent Arnoud <laurent@spkdev.net>
+Copyright (c) 2009-2021 Laurent Arnoud <laurent@spkdev.net>
 
 ---
-[![Build](https://img.shields.io/travis/spk/validate-website.svg)](https://travis-ci.org/spk/validate-website)
+[![Build](https://img.shields.io/gitlab/pipeline/spkdev/validate-website/master)](https://gitlab.com/spkdev/validate-website/-/commits/master)
+[![Coverage](https://gitlab.com/spkdev/validate-website/badges/master/coverage.svg)](https://gitlab.com/spkdev/validate-website/-/commits/master)
 [![Version](https://img.shields.io/gem/v/validate-website.svg)](https://rubygems.org/gems/validate-website)
 [![Documentation](https://img.shields.io/badge/doc-rubydoc-blue.svg)](http://www.rubydoc.info/gems/validate-website)
 [![License](https://img.shields.io/badge/license-MIT-blue.svg)](http://opensource.org/licenses/MIT "MIT")
-[![Coverage Status](https://img.shields.io/coveralls/github/spk/validate-website.svg)](https://coveralls.io/github/spk/validate-website?branch=master)
 [![Inline docs](https://inch-ci.org/github/spk/validate-website.svg?branch=master)](http://inch-ci.org/github/spk/validate-website)

data/Rakefile

@@ -1,9 +1,10 @@
+# frozen_string_literal: true
+
 require 'rake/testtask'
 require 'rubocop/rake_task'
 require 'asciidoctor'
 
-default = %i[test]
-default << :rubocop unless RUBY_ENGINE == 'rbx'
+default = %i[test rubocop]
 task default: default
 
 desc 'Update manpage from asciidoc file'

data/bin/validate-website

@@ -1,4 +1,5 @@
 #!/usr/bin/env ruby
+# frozen_string_literal: true
 
 require 'validate_website/runner'
 exit_status = ValidateWebsite::Runner.run_crawl(ARGV)

data/bin/validate-website-static

@@ -1,4 +1,5 @@
 #!/usr/bin/env ruby
+# frozen_string_literal: true
 
 require 'validate_website/runner'
 exit_status = ValidateWebsite::Runner.run_static(ARGV)

data/lib/validate_website.rb

@@ -1,2 +1,4 @@
+# frozen_string_literal: true
+
 require 'validate_website/core'
 require 'validate_website/version'

data/lib/validate_website/colorful_messages.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'paint'
 
 module ValidateWebsite
@@ -5,6 +7,7 @@ module ValidateWebsite
   module ColorfulMessages
     def color(type, message, colored = true)
       return message unless colored
+
       send(type, message)
     end
 

data/lib/validate_website/core.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'set'
 require 'open-uri'
 require 'webrick/cookie'
@@ -26,12 +28,13 @@ module ValidateWebsite
     EXIT_FAILURE_MARKUP = 64
     EXIT_FAILURE_NOT_FOUND = 65
     EXIT_FAILURE_MARKUP_NOT_FOUND = 66
-    START_MESSAGE = 'Validating'.freeze
+    START_MESSAGE = 'Validating'
 
     # Initialize core ValidateWebsite class
     # @example
     #   new({ site: "https://example.com/" }, :crawl)
     # @param [Hash] options
+    # @param [Symbol] validation_type `crawl` for web or `static` for local
     # @return [NilClass]
     def initialize(options, validation_type)
       @not_founds_count = 0
@@ -43,11 +46,11 @@ module ValidateWebsite
     end
 
     def errors?
-      @errors_count > 0
+      @errors_count.positive?
     end
 
     def not_founds?
-      @not_founds_count > 0
+      @not_founds_count.positive?
     end
 
     def exit_status
@@ -78,6 +81,7 @@ module ValidateWebsite
     def check_css_syntax(page)
       nodes = Crass::Parser.parse_stylesheet(page.body)
       return unless any_css_errors?(nodes)
+
       handle_validation_error(page.url)
     end
 
@@ -111,10 +115,10 @@ module ValidateWebsite
     # @param [Nokogiri::HTML::Document] original_doc
     # @param [String] The raw HTTP response body of the page
     # @param [String] url
-    # @param [Regexp] Errors to ignore
+    # @param [Hash] Validator options
     #
-    def validate(doc, body, url, ignore = nil)
-      validator = Validator.new(doc, body, ignore: ignore)
+    def validate(doc, body, url, options)
+      validator = Validator.new(doc, body, **options)
       if validator.valid?
         print color(:success, '.', options[:color]) # rspec style
       else
@@ -125,6 +129,7 @@ module ValidateWebsite
     def handle_html_validation_error(validator, url)
       handle_validation_error(url)
       return unless options[:verbose]
+
       puts color(:error, validator.errors.join(', '), options[:color])
     end
 

data/lib/validate_website/crawl.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'validate_website/core'
 require 'validate_website/utils'
 
@@ -41,9 +43,10 @@ module ValidateWebsite
     #
     def extract_imgs_from_page(page)
       return Set[] if page.is_redirect?
+
       page.doc.search('//img[@src]').reduce(Set[]) do |result, elem|
         u = elem.attributes['src'].content
-        result << page.to_absolute(URI.parse(URI.encode(u)))
+        result << page.to_absolute(URI.parse(WEBrick::HTTPUtils.escape(u)))
       end
     end
 
@@ -77,7 +80,10 @@ module ValidateWebsite
         end
 
         if validate?(page)
-          validate(page.doc, page.body, page.url, options[:ignore])
+          keys = %i[ignore html5_validator]
+          # slice does not exists on Ruby <= 2.4
+          slice = Hash[[keys, options.values_at(*keys)].transpose]
+          validate(page.doc, page.body, page.url, slice)
         end
       end
     end

data/lib/validate_website/option_parser.rb

@@ -1,4 +1,7 @@
+# frozen_string_literal: true
+
 require 'slop'
+require File.expand_path('version', __dir__)
 
 module ValidateWebsite
   # Internal class for parse command line args
@@ -28,6 +31,7 @@ module ValidateWebsite
     # Generic parse method for crawl or static options
     def self.parse(options, type)
       raise ArgumentError unless VALID_TYPES.include?(type)
+
       # We are in command line (ARGV)
       if options.is_a?(Array)
         send("command_line_parse_#{type}", options)
@@ -52,7 +56,7 @@ module ValidateWebsite
       opt.regexp('-i', '--ignore',
                  'Validation errors to ignore (ex: "valign|autocorrect")')
       opt.string('-x', '--html5-validator',
-                 'Change default html5 validator engine (ex: tidy or nu)',
+                 'Change default html5 validator engine (tidy/nu/nokogumbo)',
                  default: DEFAULT_OPTIONS[:html5_validator])
       opt.string('-5', '--html5-validator-service-url',
                  'Change default html5 validator service URL for "nu" engine')

data/lib/validate_website/runner.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'validate_website/core'
 
 module ValidateWebsite
@@ -5,7 +7,7 @@ module ValidateWebsite
   class Runner
     def self.trap_interrupt
       trap('INT') do
-        STDERR.puts "\nExiting..."
+        warn "\nExiting..."
         exit!(1)
       end
     end

data/lib/validate_website/static.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'validate_website/core'
 require 'validate_website/utils'
 
@@ -5,7 +7,7 @@ module ValidateWebsite
   # Class for validation Static website
   class Static < Core
     CONTENT_TYPES = ['text/html', 'text/xhtml+xml'].freeze
-    START_MESSAGE_TYPE = 'files'.freeze
+    START_MESSAGE_TYPE = 'files'
 
     attr_reader :history_count
 
@@ -24,7 +26,8 @@ module ValidateWebsite
       files = Dir.glob(@options[:pattern])
       files.each do |file|
         next unless File.file?(file)
-        next if @options[:exclude] && @options[:exclude].match(file)
+        next if @options[:exclude]&.match(file)
+
         @history_count += 1
         check_static_file(file)
       end
@@ -57,7 +60,10 @@ module ValidateWebsite
 
     def check_page(file, page)
       if page.html? && options[:markup]
-        validate(page.doc, page.body, file, options[:ignore])
+        keys = %i[ignore html5_validator]
+        # slice does not exists on Ruby <= 2.4
+        slice = Hash[[keys, options.values_at(*keys)].transpose]
+        validate(page.doc, page.body, file, slice)
       end
       check_static_not_found(page.links) if options[:not_found]
     end
@@ -68,11 +74,13 @@ module ValidateWebsite
       static_links = links.map { |l| StaticLink.new(l, @site) }
       static_links.each do |static_link|
         next unless static_link.check?
+
         unless File.exist?(static_link.file_path)
           not_found_error(static_link.file_path)
           next
         end
         next unless static_link.css?
+
         check_static_not_found static_link.extract_urls_from_fake_css_response
       end
     end

data/lib/validate_website/static_link.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'uri'
 require 'validate_website/utils'
 require 'validate_website/static'
@@ -6,7 +8,7 @@ require 'spidr'
 # rubocop:disable Metrics/BlockLength
 StaticLink = Struct.new(:link, :site) do
   def link_uri
-    @link_uri = URI.parse(URI.encode(link))
+    @link_uri = URI.parse(WEBrick::HTTPUtils.escape(link))
     @link_uri = URI.join(site, @link_uri) if @link_uri.host.nil?
     @link_uri
   end
@@ -25,9 +27,9 @@ StaticLink = Struct.new(:link, :site) do
 
   def body
     if File.exist?(link)
-      open(link).read
+      File.open(link).read
     else
-      open(file_path).read
+      File.open(file_path).read
     end
   end
 

data/lib/validate_website/utils.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 # Base module ValidateWebsite
 module ValidateWebsite
   # Utils class for CSS helpers
@@ -10,6 +12,7 @@ module ValidateWebsite
     def self.extract_urls_from_css(page)
       return Set[] unless page
       return Set[] if page.body.nil?
+
       nodes = Crass::Parser.parse_stylesheet(page.body)
       extract_urls_from_nodes nodes, page
     end

data/lib/validate_website/validator.rb

@@ -1,6 +1,8 @@
+# frozen_string_literal: true
+
 require 'uri'
 
-require 'nokogiri'
+require 'nokogumbo' unless ValidateWebsite.jruby?
 require 'w3c_validators'
 
 require 'validate_website/validator_class_methods'
@@ -10,27 +12,25 @@ module ValidateWebsite
   class Validator
     extend ValidatorClassMethods
 
-    @html5_validator_service_url = 'https://checker.html5.org/'
+    @html5_validator_service_url = 'https://validator.nu/'
+    XHTML_SCHEMA_PATH = File.expand_path('../../data/schemas', __dir__)
+    @mutex = Mutex.new
+
     class << self
       attr_accessor :html5_validator_service_url
-    end
-
-    XHTML_PATH = File.expand_path('../../data/schemas', __dir__)
 
-    @xsd_schemas ||= {}
-
-    # `Dir.chdir` is needed by `Nokogiri::XML::Schema` to validate with local
-    # files and cannot use file absolute path.
-    Dir.glob(File.join(XHTML_PATH, '*.xsd')).each do |schema|
-      Dir.chdir(XHTML_PATH) do
-        schema_name = File.basename(schema, '.xsd')
-        schema_content = File.read(File.basename(schema))
-        begin
-          @xsd_schemas[schema_name] = Nokogiri::XML::Schema(schema_content)
-        rescue Nokogiri::XML::SyntaxError
-          STDERR.puts "XSD SCHEMA: #{schema} cannot be loaded"
+      # http://www.w3.org/TR/xhtml1-schema/
+      def schema(namespace)
+        @mutex.synchronize do
+          Dir.chdir(XHTML_SCHEMA_PATH) do
+            if File.exist?("#{namespace}.xsd")
+              Nokogiri::XML::Schema(File.read("#{namespace}.xsd"))
+            end
+          end
         end
       end
+
+      alias xsd schema
     end
 
     attr_reader :original_doc, :body, :dtd, :doc, :namespace, :html5_validator
@@ -65,25 +65,22 @@ module ValidateWebsite
       @ignore ? @errors.reject { |e| @ignore =~ e } : @errors
     end
 
-    # http://www.w3.org/TR/xhtml1-schema/
-    def self.xsd(namespace)
-      return unless namespace
-      @xsd_schemas[namespace] if @xsd_schemas.key? namespace
-    end
-
     private
 
     # http://www.w3.org/TR/xhtml1/#dtds
     def find_namespace(dtd)
       return unless dtd.system_id
+
       dtd_uri = URI.parse(dtd.system_id)
       return unless dtd_uri.path
+
       @dtd_uri = dtd_uri
       File.basename(@dtd_uri.path, '.dtd')
     end
 
     def document
       return @document if @document
+
       @document = if @dtd_uri && @body.match(@dtd_uri.to_s)
                     @body.sub(@dtd_uri.to_s, @namespace + '.dtd')
                   else
@@ -95,11 +92,11 @@ module ValidateWebsite
     def validate
       if document =~ /^\<!DOCTYPE html\>/i
         html5_validate
-      elsif self.class.xsd(@namespace)
-        self.class.xsd(@namespace).validate(xhtml_doc)
+      elsif self.class.schema(@namespace)
+        self.class.schema(@namespace).validate(xhtml_doc)
       else
         # dont have xsd fall back to dtd
-        Dir.chdir(XHTML_PATH) do
+        Dir.chdir(XHTML_SCHEMA_PATH) do
           Nokogiri::HTML.parse(document)
         end.errors
       end
@@ -115,8 +112,10 @@ module ValidateWebsite
     def html5_validate
       if html5_validator.to_sym == :tidy && self.class.tidy
         tidy_validate
-      else
+      elsif html5_validator.to_sym == :nu
         nu_validate
+      else
+        Nokogiri::HTML5(document, max_errors: -1).errors
       end
     end
 
@@ -138,7 +137,7 @@ module ValidateWebsite
     end
 
     def xhtml_doc
-      Dir.chdir(XHTML_PATH) do
+      Dir.chdir(XHTML_SCHEMA_PATH) do
         Nokogiri::XML(document) { |cfg| cfg.nonoent.dtdload.dtdvalid.nonet }
       end
     end