gherkin_language 0.0.7 → 0.0.8

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 04aee53e8b36b777bf0889ce6b0cdcd94fdef265
-  data.tar.gz: 54aaa1ba0d8625f05aeb8ee79e09878657988957
+  metadata.gz: db5be66787bff52612008a4e2999084191cdff53
+  data.tar.gz: 294d5c8d58ea483b2bb877f90cdbe4e7a0bae614
 SHA512:
-  metadata.gz: b3186581735b4460df28c170f865f4ee3cbb4e203cba63d77e78c3d6e1d9622a777483b9e104749972847c1a7eca36f93cd89c7f46afad1fca6be17d42903f12
-  data.tar.gz: 4563bb4bdfe7815bf4573c8352160c6db8fb86d01a120c0ae1155a053f7fdd423cc5e012f0a607f9d7ac17b98019c36a123147ae51d6571c555b863ad2cc0dc1
+  metadata.gz: 65eb341ef2cf91a80d11c42f531c20a1dc994e751dd0b538f579efb14a8065ef5acdb41ff7aedf7fc903a6c528b7261a00b30b6c7e9064861c004fd52aac2790
+  data.tar.gz: e3916bf3b624a8b8564122e9d4d05f16cbd1cee536d77e0754e88669377519fcdfbb27b507698bb63ae3e37cc3fc334cdc28360623a4274ee90862c466a70cc2

data/.rubocop.yml

@@ -7,21 +7,21 @@
 
 # Offense count: 5
 Metrics/AbcSize:
-  Max: 48
+  Max: 36
 
 # Offense count: 2
 # Configuration parameters: CountComments.
 Metrics/ClassLength:
-  Max: 171
+  Max: 172
 
 # Offense count: 2
 Metrics/CyclomaticComplexity:
-  Max: 11
+  Max: 7
 
 # Offense count: 17
 # Configuration parameters: AllowURI, URISchemes.
 Metrics/LineLength:
-  Max: 121
+  Max: 117
 
 # Offense count: 8
 # Configuration parameters: CountComments.
@@ -34,5 +34,5 @@ Metrics/ParameterLists:
   Max: 8
 
 # Offense count: 2
-Metrics/PerceivedComplexity:
-  Max: 12
+#Metrics/PerceivedComplexity:
+#  Max: 8

data/README.md

@@ -23,6 +23,8 @@ To tag all words used, start it with `--tag`. This allows to build up a glossary
 
 To ignore specific rules, mention them with an `--ignore RULE`. This allows to bypass the checks.
 
+To check for confused words, based on ngrams, add `--ngram`. Please note, that it requires much disk space and time.
+
 
 Glossary
 --------

data/Rakefile

@@ -18,7 +18,7 @@ task test: :cucumber
 
 desc 'Publishes the Gem'
 task push: :build do
-  sh 'gem push gherkin_language-0.0.7.gem'
+  sh 'gem push gherkin_language-0.0.8.gem'
 end
 
 desc 'Checks ruby style'

data/bin/gherkin_language

@@ -17,9 +17,12 @@ OptionParser.new do |opts|
   opts.on('--no-cache', 'do not use cache') do |no_cache|
     options[:no_cache] = no_cache
   end
+  opts.on('--ngram', 'check based on ngrams') do |ngram|
+    options[:ngram] = ngram
+  end
 end.parse!
 
-language = GherkinLanguage.new(options.key? :no_cache)
+language = GherkinLanguage.new(options.key?(:no_cache), options.key?(:ngram))
 
 if options.key? :sentences
   ARGV.each { |file| puts language.extract_sentences language.parse file }

data/features/confused_words.feature

@@ -0,0 +1,52 @@
+Feature: Confused Words
+  As a Business Analyst
+  I want to be informed for confused words
+  so that I know when I'm using wrong words
+
+  Background:
+    Given a file named "confused_words.rb" with:
+      """
+      $LOAD_PATH << '../../lib'
+      require 'gherkin_language'
+
+      no_cache = true
+      ngrams = true
+      language = GherkinLanguage.new(no_cache, ngrams)
+      language.analyze 'test.feature'
+      exit language.report
+
+      """
+
+  Scenario: Warns for confused word
+    Given a file named "test.feature" with:
+      """
+      Feature: Test
+        Scenario: Tag
+          Given I do not now where it is
+          When execute
+          Then pass
+      """
+    When I run `ruby confused_words.rb`
+    Then it should fail with exactly:
+      """
+      [non-conformance] CONFUSION_RULE
+        Statistic suggests that 'know' (to be aware of) might be the correct word here, not 'now' (in this moment). Please check.
+        Context: Given I do not now where it is when execute then pass
+        Replacements: know
+        References: test.feature
+      """
+
+  Scenario: Accept non confused words
+    Given a file named "test.feature" with:
+      """
+      Feature: Test
+        Scenario: Tag
+          Given I do not know where it is
+          When execute
+          Then pass
+      """
+    When I run `ruby confused_words.rb`
+    Then it should pass with exactly:
+      """
+
+      """

data/features/sentences.feature

@@ -80,18 +80,18 @@ Feature: Sentences
       """
       Feature: Test
         Scenario Outline: Test
-          When <A> <B>
-          Then <C>
+          When <A>
+          Then <B>
 
           Examples: Table
-            | A |
-            | 1 |
-            | 2 |
+            | A | B |
+            | 1 | 3 |
+            | 2 | 4 |
 
           Examples: Second Table
-            | B | C |
-            | 3 | 5 |
-            | 4 | 6 |
+            | B | A |
+            | 5 | 7 |
+            | 6 | 8 |
       """
     When I run `ruby extract_sentences.rb`
     Then it should pass with:
@@ -99,11 +99,11 @@ Feature: Sentences
       Test
       Test
       Table
+      When 1 then 3
+      When 2 then 4
       Second Table
-      When 1 3 then 5
-      When 1 4 then 6
-      When 2 3 then 5
-      When 2 4 then 6
+      When 7 then 5
+      When 8 then 6
       """
 
   Scenario: Extract Sentences considers feature description

data/gherkin_language.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name        = 'gherkin_language'
-  s.version     = '0.0.7'
+  s.version     = '0.0.8'
   s.date        = '2015-10-04'
   s.summary     = 'Gherkin Language'
   s.description = 'Check language of Gherkin Files'

data/lib/gherkin_language.rb

@@ -1,6 +1,8 @@
 # encoding: utf-8
 require 'gherkin/formatter/json_formatter'
 require 'gherkin/parser/parser'
+require 'gherkin_language/error'
+require 'gherkin_language/language_tool_process'
 require 'rexml/document'
 require 'stringio'
 require 'multi_json'
@@ -14,143 +16,7 @@ require 'digest'
 
 # gherkin utilities
 class GherkinLanguage
-  # This service class provides access to language tool process.
-  class LanguageToolProcess
-    attr_accessor :errors, :unknown_words
-
-    VERSION = 'LanguageTool-3.0'
-    URL = "https://www.languagetool.org/download/#{VERSION}.zip"
-
-    # This value entity class represents a language error
-    class Error
-      attr_accessor :category, :context, :issuetype, :message, :replacements, :rule, :from_y, :to_y
-
-      def initialize(category, context, issuetype, message, replacements, rule, from_y, to_y)
-        @category = category
-        @context = context
-        @issuetype = issuetype
-        @message = message
-        @replacements = replacements
-        @rule = rule
-        @from_y = from_y
-        @to_y = to_y
-      end
-
-      def str(references)
-        (red("[#{@issuetype}] #{@rule}\n") +
-         "  #{@message}\n  Context: #{@context}\n  Replacements: #{@replacements}\n  References: #{references * ', '}\n")
-      end
-    end
-
-    def initialize
-      path = Dir.tmpdir
-      download path unless File.exist? "#{path}/#{VERSION}/languagetool-commandline.jar"
-      @path = path
-      @p = nil
-      @reference_line = 0
-      @errors = []
-      @unknown_words = []
-      use_user_glossary "#{path}/#{VERSION}" if File.exist? '.glossary'
-    end
-
-    def use_user_glossary(path)
-      resource_path = "#{path}/org/languagetool/resource/en"
-      system "cp #{resource_path}/added.txt #{resource_path}/added.copy && cp .glossary #{resource_path}/added.txt"
-      at_exit do
-        system "cp #{resource_path}/added.copy #{resource_path}/added.txt"
-      end
-    end
-
-    def download(path)
-      system "wget --quiet #{URL} -O /var/tmp/languagetool.zip"
-      FileUtils.mkdir_p path
-      system "unzip -qq -u /var/tmp/languagetool.zip -d #{path}"
-    end
-
-    def start!
-      @errors = []
-      @unknown_words = []
-      @reference_line = 0
-      Dir.chdir("#{@path}/#{VERSION}/") do
-        @p = IO.popen('java -jar languagetool-commandline.jar --list-unknown --api --language en-US -', 'r+')
-      end
-    end
-
-    def tag(sentences)
-      output = ''
-      Dir.chdir("#{@path}/#{VERSION}/") do
-        p = IO.popen('java -jar languagetool-commandline.jar --taggeronly --api --language en-US -', 'r+')
-        sentences.each { |sentence| p.write sentence }
-        p.close_write
-        line = p.readline
-        loop do
-          break if line == "<!--\n"
-          output << line
-          line = p.readline
-        end
-        p.close
-      end
-      output.gsub!(' ', "\n")
-      output.gsub!(']', "]\n")
-      output.gsub!("\n\n", "\n")
-      output
-    end
-
-    def check_paragraph(paragraph)
-      start_line = @reference_line
-      send paragraph
-      end_line = @reference_line
-      send "\n\n"
-      Range.new(start_line, end_line)
-    end
-
-    def send(sentence)
-      @reference_line += sentence.count "\n"
-      @p.write sentence
-    end
-
-    def parse_errors(result)
-      doc = REXML::Document.new result
-      errors = []
-      doc.elements.each '//error' do |error|
-        errors.push Error.new(
-          error.attributes['category'],
-          error.attributes['context'].strip,
-          error.attributes['locqualityissuetype'],
-          error.attributes['msg'],
-          error.attributes['replacements'],
-          error.attributes['ruleId'],
-          error.attributes['fromy'].to_i,
-          error.attributes['toy'].to_i)
-      end
-      errors
-    end
-
-    def parse_unknown_words(result)
-      doc = REXML::Document.new result
-      errors = []
-      doc.elements.each '//unknown_words/word' do |error|
-        errors.push error.text
-      end
-      errors
-    end
-
-    def stop!
-      @p.close_write
-      errors = ''
-      line = @p.readline
-      loop do
-        break if line == "<!--\n"
-        errors << line
-        line = @p.readline
-      end
-      @errors = parse_errors errors
-      @unknown_words = parse_unknown_words errors
-      @p.close
-    end
-  end
-
-  def initialize(no_cache = false)
+  def initialize(no_cache = false, ngram = false)
     path = "~/.gherkin_language/#{LanguageToolProcess::VERSION}/accepted_paragraphs.yml"
     @settings_path = File.expand_path path
     @accepted_paragraphs = {}
@@ -162,6 +28,7 @@ class GherkinLanguage
     @references = {}
     @line_to_reference = {}
     @exceptions = []
+    @ngram = ngram
   end
 
   def ignore(exception)
@@ -219,7 +86,7 @@ class GherkinLanguage
 
   def report
     return 0 if @references.keys.empty?
-    language = LanguageToolProcess.new
+    language = LanguageToolProcess.new @ngram
     language.start!
 
     @references.keys.each do |sentence|
@@ -294,43 +161,47 @@ class GherkinLanguage
     background = []
     input.each do |features|
       next unless features.key? 'elements'
-      elements = features['elements']
-      elements.each do |scenario|
+      features['elements'].each do |scenario|
         next unless scenario.key? 'steps'
         terms = background.dup
         if scenario['type'] == 'background'
-          scenario['steps'].each do |step|
-            new_terms = [step['keyword'], step['name']].join
-            new_terms = uncapitalize(new_terms) unless terms.empty?
-            background.push new_terms
-          end
+          background.push extract_terms_from_scenario(scenario['steps'], terms)
           next
         end
 
-        scenario['steps'].each do |step|
-          keyword = step['keyword']
-          keyword = 'and ' unless background.empty? || keyword != 'Given '
-          new_terms = [keyword, step['name']].join
-          new_terms = uncapitalize(new_terms) unless terms.empty?
-          terms.push new_terms
-        end
-        sentence = terms.join ' '
+        terms.push extract_terms_from_scenario(scenario['steps'], background)
+        sentence = terms.join(' ').strip
         if scenario.key? 'examples'
-          prototypes = [sentence.strip]
-          scenario['examples'].each do |example|
-            sentences.push example['name'] unless example['name'].empty?
-            sentences.push example['description'] unless example['description'].empty?
-            prototypes = prototypes.map { |prototype| expand_outlines(prototype, example) }.flatten
-          end
-          sentences += prototypes
+          sentences += extract_examples(scenario['examples'], sentence)
         else
-          sentences.push sentence.strip
+          sentences.push sentence
         end
       end
     end
     sentences
   end
 
+  def extract_terms_from_scenario(steps, background)
+    steps.map do |step|
+      keyword = step['keyword']
+      keyword = 'and ' unless background.empty? || keyword != 'Given '
+      terms = [keyword, step['name']].join
+      terms = uncapitalize(terms) unless background.empty?
+      background = terms
+      terms
+    end.flatten
+  end
+
+  def extract_examples(examples, prototype)
+    examples.map do |example|
+      sentences = []
+      sentences.push example['name'] unless example['name'].empty?
+      sentences.push example['description'] unless example['description'].empty?
+      sentences += expand_outlines(prototype, example)
+      sentences
+    end.flatten
+  end
+
   def uncapitalize(term)
     term[0, 1].downcase + term[1..-1]
   end

data/lib/gherkin_language/error.rb

@@ -0,0 +1,34 @@
+# encoding: utf-8
+require 'gherkin/formatter/json_formatter'
+require 'gherkin/parser/parser'
+require 'rexml/document'
+require 'stringio'
+require 'multi_json'
+require 'term/ansicolor'
+include Term::ANSIColor
+require 'tmpdir'
+require 'fileutils'
+require 'yaml'
+require 'set'
+require 'digest'
+
+# This value entity class represents a language error
+class Error
+  attr_accessor :category, :context, :issuetype, :message, :replacements, :rule, :from_y, :to_y
+
+  def initialize(category, context, issuetype, message, replacements, rule, from_y, to_y)
+    @category = category
+    @context = context
+    @issuetype = issuetype
+    @message = message
+    @replacements = replacements
+    @rule = rule
+    @from_y = from_y
+    @to_y = to_y
+  end
+
+  def str(references)
+    (red("[#{@issuetype}] #{@rule}\n") +
+     "  #{@message}\n  Context: #{@context}\n  Replacements: #{@replacements}\n  References: #{references * ', '}\n")
+  end
+end

data/lib/gherkin_language/language_tool_process.rb

@@ -0,0 +1,142 @@
+# encoding: utf-8
+require 'gherkin/formatter/json_formatter'
+require 'gherkin/parser/parser'
+require 'rexml/document'
+require 'stringio'
+require 'multi_json'
+require 'term/ansicolor'
+include Term::ANSIColor
+require 'tmpdir'
+require 'fileutils'
+require 'yaml'
+require 'set'
+require 'digest'
+
+# This service class provides access to language tool process.
+class LanguageToolProcess
+  attr_accessor :errors, :unknown_words
+
+  VERSION = 'LanguageTool-3.1'
+  URL = "https://www.languagetool.org/download/#{VERSION}.zip"
+  NGRAM_VERSION = 'ngrams-en-20150817'
+  NGRAM_URL = "https://languagetool.org/download/ngram-data/#{NGRAM_VERSION}.zip"
+
+  def initialize(ngrams = false)
+    path = Dir.tmpdir
+    download(path, URL) unless File.exist? "#{path}/#{VERSION}/languagetool-commandline.jar"
+    if ngrams
+      @ngrams_path = "#{path}/#{NGRAM_VERSION}"
+      download("#{@ngrams_path}/en", NGRAM_URL) unless File.exist? @ngrams_path
+    end
+    @path = path
+    @p = nil
+    @reference_line = 0
+    @errors = []
+    @unknown_words = []
+    @ngrams = ngrams
+    use_user_glossary "#{path}/#{VERSION}" if File.exist? '.glossary'
+  end
+
+  def use_user_glossary(path)
+    resource_path = "#{path}/org/languagetool/resource/en"
+    system "cp #{resource_path}/added.txt #{resource_path}/added.copy && cp .glossary #{resource_path}/added.txt"
+    at_exit do
+      system "cp #{resource_path}/added.copy #{resource_path}/added.txt"
+    end
+  end
+
+  def download(path, url)
+    system "wget --quiet #{url} -O /var/tmp/languagetool.zip"
+    FileUtils.mkdir_p path
+    system "unzip -qq -u /var/tmp/languagetool.zip -d #{path}"
+    system 'rm /var/tmp/languagetool.zip'
+  end
+
+  def start!
+    @errors = []
+    @unknown_words = []
+    @reference_line = 0
+    Dir.chdir("#{@path}/#{VERSION}/") do
+      command = 'java -jar languagetool-commandline.jar --list-unknown --api --language en-US'
+      command += " --languagemodel #{@ngrams_path}" if @ngrams
+      @p = IO.popen("#{command} -", 'r+')
+    end
+  end
+
+  def tag(sentences)
+    output = ''
+    Dir.chdir("#{@path}/#{VERSION}/") do
+      p = IO.popen('java -jar languagetool-commandline.jar --taggeronly --api --language en-US -', 'r+')
+      sentences.each { |sentence| p.write sentence }
+      p.close_write
+      line = p.readline
+      loop do
+        break if line == "<!--\n"
+        output << line
+        line = p.readline
+      end
+      p.close
+    end
+    output.gsub!(' ', "\n")
+    output.gsub!(']', "]\n")
+    output.gsub!("\n\n", "\n")
+    output
+  end
+
+  def check_paragraph(paragraph)
+    start_line = @reference_line
+    send paragraph
+    end_line = @reference_line
+    send "\n\n"
+    Range.new(start_line, end_line)
+  end
+
+  def send(sentence)
+    @reference_line += sentence.count "\n"
+    @p.write sentence
+  end
+
+  def parse_errors(result)
+    doc = REXML::Document.new result
+    errors = []
+    doc.elements.each '//error' do |error|
+      errors.push decode_error error
+    end
+    errors
+  end
+
+  def decode_error(error)
+    Error.new(
+      error.attributes['category'],
+      error.attributes['context'].strip,
+      error.attributes['locqualityissuetype'],
+      error.attributes['msg'],
+      error.attributes['replacements'],
+      error.attributes['ruleId'],
+      error.attributes['fromy'].to_i,
+      error.attributes['toy'].to_i)
+  end
+
+  def parse_unknown_words(result)
+    doc = REXML::Document.new result
+    errors = []
+    doc.elements.each '//unknown_words/word' do |error|
+      errors.push error.text
+    end
+    errors
+  end
+
+  def stop!
+    @p.close_write
+    errors = ''
+    line = @p.readline
+    loop do
+      break if line == "<!--\n"
+      errors << line
+      line = @p.readline
+    end
+    @errors = parse_errors errors
+    @unknown_words = parse_unknown_words errors
+    @p.close
+  end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: gherkin_language
 version: !ruby/object:Gem::Version
-  version: 0.0.7
+  version: 0.0.8
 platform: ruby
 authors:
 - Stefan Rohe
@@ -67,6 +67,7 @@ files:
 - README.md
 - Rakefile
 - bin/gherkin_language
+- features/confused_words.feature
 - features/correct.feature
 - features/exception.feature
 - features/glossary.feature
@@ -76,6 +77,8 @@ files:
 - features/tag.feature
 - gherkin_language.gemspec
 - lib/gherkin_language.rb
+- lib/gherkin_language/error.rb
+- lib/gherkin_language/language_tool_process.rb
 - test/test_gherkin_language.rb
 - test/test_gherkin_language_tool.rb
 homepage: http://github.com/funkwerk/gherkin_language/