gherkin_language 0.0.2

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA1:
+  metadata.gz: 6e36578ba3f6c90fdc22f8ddde30d96a0ac9c955
+  data.tar.gz: 96d5cfa9266ad79f3c287c7101b27669cf47e009
+SHA512:
+  metadata.gz: 550d087efcbff10a081dda79bab4d26060c6c84df5777a1ae24112dbafbd61ca11775685bf43f7b1b8a84d8e23670a307d907467863be3fc9181ecad4b60804b
+  data.tar.gz: 951757d91180b2cdb0720ecb1e325aab15779e34e86a532b3222b63a30effe5a1ee7987c5070fccd510da2adb6b3321de3564498627532a85a31e84b4b237dda

data/.rubocop.yml

@@ -0,0 +1,38 @@
+# This configuration was generated by `rubocop --auto-gen-config`
+# on 2015-07-02 21:10:32 +0200 using RuboCop version 0.31.0.
+# The point is for the user to remove these configuration records
+# one by one as the offenses are removed from the code base.
+# Note that changes in the inspected code, or installation of new
+# versions of RuboCop, may require this file to be generated again.
+
+# Offense count: 5
+Metrics/AbcSize:
+  Max: 47
+
+# Offense count: 2
+# Configuration parameters: CountComments.
+Metrics/ClassLength:
+  Max: 159
+
+# Offense count: 2
+Metrics/CyclomaticComplexity:
+  Max: 11
+
+# Offense count: 17
+# Configuration parameters: AllowURI, URISchemes.
+Metrics/LineLength:
+  Max: 121
+
+# Offense count: 8
+# Configuration parameters: CountComments.
+Metrics/MethodLength:
+  Max: 36
+
+# Offense count: 1
+# Configuration parameters: CountKeywordArgs.
+Metrics/ParameterLists:
+  Max: 8
+
+# Offense count: 2
+Metrics/PerceivedComplexity:
+  Max: 12

data/.travis.yml

@@ -0,0 +1,4 @@
+language: ruby
+rvm:
+ - 2.1.0
+ - 2.0.0

data/Gemfile

@@ -0,0 +1,6 @@
+source 'https://rubygems.org'
+gem 'rubocop'
+gem 'gherkin'
+gem 'rake'
+gem 'aruba'
+gem 'term-ansicolor'

data/Guardfile

@@ -0,0 +1,3 @@
+guard 'rake', task: :default do
+  watch(%r{^src/.*$})
+end

data/LICENSE

@@ -0,0 +1,22 @@
+The MIT License (MIT)
+
+Copyright (c) 2015 Funkwerk AG
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
+

data/README.md

@@ -0,0 +1,29 @@
+Check Language of Gherkin Files
+===============================
+
+[![Build Status](https://travis-ci.org/funkwerk/gherkin_language.svg)](https://travis-ci.org/funkwerk/gherkin_language)
+
+This tool analyzes the language of gherkin files and report language errors.
+Currently just English is supported.
+
+Usage
+-----
+
+run `gherkin_language` on a list of files
+
+    gherkin_language FEATURE_FILES
+
+`gherkin_language` caches valid sentences. If all sentences are valid, it just needs some time to hash them.
+
+To disable usage of cache, start it with `--no-cache`.
+
+To just extract the sentences, start it with `--sentences`. This could be helpful for using these sentences in another tool. It should not be used for formatting issues. For formatting `gherkin_format --template ...` could be used.
+
+To tag all words used, start it with `--tag`. This allows to build up a glossary based on the feature files provided.
+
+
+Glossary
+--------
+
+It happens that there are words which are unknown to the dictionary.
+Once this happens think about if could use another word, that is more common. If there is no such word, add it to the directory-located glossary.

data/Rakefile

@@ -0,0 +1,29 @@
+require 'rake/testtask'
+
+task default: :build
+
+desc 'Builds the Gem.'
+task build: :test do
+  sh 'gem build gherkin_language.gemspec'
+end
+
+Rake::TestTask.new do |t|
+  t.libs << 'test'
+end
+
+task test: :rubocop
+task test: :cucumber
+
+desc 'Publishes the Gem'
+task :push do
+  sh 'gem push gherkin_language-0.0.2.gem'
+end
+
+desc 'Checks ruby style'
+task :rubocop do
+  sh 'rubocop'
+end
+
+task :cucumber do
+  sh 'cucumber'
+end

data/bin/gherkin_language

@@ -0,0 +1,34 @@
+#!/usr/bin/env ruby
+require 'gherkin_language'
+require 'optparse'
+
+options = {}
+OptionParser.new do |opts|
+  opts.banner = 'Usage: gherkin_language [files]'
+  opts.on('-v', '--[no-]verbose', 'Run verbosely') do |verbose|
+    options[:verbose] = verbose
+  end
+  opts.on('--sentences', 'extract relevant sentences') do |sentences|
+    options[:sentences] = sentences
+  end
+  opts.on('--tag', 'tag words') do |tag|
+    options[:tag] = tag
+  end
+  opts.on('--no-cache', 'do not use cache') do |no_cache|
+    options[:no_cache] = no_cache
+  end
+end.parse!
+
+language = GherkinLanguage.new(options.key? :no_cache)
+
+if options.key? :sentences
+  ARGV.each { |file| puts language.extract_sentences language.parse file }
+  exit
+end
+if options.key? :tag
+  puts language.tag ARGV
+  exit
+end
+
+ARGV.each { |file| language.analyze file }
+exit language.report

data/features/report.feature

@@ -0,0 +1,58 @@
+Feature: Report
+  As a Business Analyst
+  I want to get a report for language mistakes within my feature files
+  so that I am able to fix them
+
+  Background:
+    Given a file named "report.rb" with:
+      """
+      $LOAD_PATH << '../../lib'
+      require 'gherkin_language'
+
+      no_cache = true
+      language = GherkinLanguage.new no_cache
+      language.analyze 'test.feature'
+      exit language.report
+
+      """
+
+  Scenario: Broken Sentence
+    Given a file named "test.feature" with:
+      """
+      Feature: Test
+        Scenario: Tag
+          Given an test
+          When execute
+          Then pass
+      """
+    When I run `ruby report.rb`
+    Then it should fail with exactly:
+      """
+      [misspelling] EN_A_VS_AN
+        Use 'a' instead of 'an' if the following word doesn't start with a vowel sound, e.g. 'a sentence', 'a university'
+        Context: Given an test when execute then pass  
+        Replacements: a
+        References: test.feature
+
+      """
+
+  Scenario: Unknown Words
+    Given a file named "test.feature" with:
+      """
+      Feature: Test
+        Scenario: Tag
+          Given a test
+          When exicute
+          Then pass
+      """
+    When I run `ruby report.rb`
+    Then it should fail with exactly:
+      """
+      [misspelling] MORFOLOGIK_RULE_EN_US
+        Possible spelling mistake found
+        Context: Given a test when exicute then pass  
+        Replacements: execute
+        References: test.feature
+      1 unknown words: exicute
+
+      """

data/features/sentences.feature

@@ -0,0 +1,127 @@
+Feature: Sentences
+  As a Business Analyst
+  I want to extract all sentences out of my feature files
+  so that I could do further analysis on them
+
+  Background:
+    Given a file named "extract_sentences.rb" with:
+      """
+      $LOAD_PATH << '../../lib'
+      require 'gherkin_language'
+
+      language = GherkinLanguage.new false
+      puts language.extract_sentences language.parse 'foo.feature'
+
+      """
+
+  Scenario: Extract Sentences
+    Given a file named "foo.feature" with:
+      """
+      Feature: Foo
+        Scenario: Bar
+          Given a foo
+          When I bar
+          Then I baz
+      """
+    When I run `ruby extract_sentences.rb`
+    Then it should pass with:
+      """
+      Foo
+      Bar
+      Given a foo when I bar then I baz
+      """
+
+  Scenario: Extract Sentences with background
+    Given a file named "foo.feature" with:
+      """
+      Feature: Foo
+        Background:
+          Given something
+
+        Scenario: Bar
+          Given a foo
+          When I bar
+          Then I baz
+      """
+    When I run `ruby extract_sentences.rb`
+    Then it should pass with:
+      """
+      Foo
+      Bar
+      Given something and a foo when I bar then I baz
+      """
+
+  Scenario: Extract Sentences from outlines
+    Given a file named "foo.feature" with:
+      """
+      Feature: Foo
+        Scenario Outline: Bar
+          Given a <foo>
+          When I <bar>
+          Then I <baz>
+
+          Examples: table
+            | foo | bar | baz |
+            | FOO | BAR | BAZ |
+            | oof | rab | zab |
+      """
+    When I run `ruby extract_sentences.rb`
+    Then it should pass with:
+      """
+      Foo
+      Bar
+      table
+      Given a FOO when I BAR then I BAZ
+      Given a oof when I rab then I zab
+      """
+
+  Scenario: Extract Sentences considers feature description
+    Given a file named "foo.feature" with:
+      """
+      Feature: Foo
+        As a user,
+        I want something
+        so that I have that
+
+        Scenario: Bar
+          Given a foo
+          When I bar
+          Then I baz
+      """
+    When I run `ruby extract_sentences.rb`
+    Then it should pass with:
+      """
+      Foo
+      As a user,
+      I want something
+      so that I have that
+      Bar
+      Given a foo when I bar then I baz
+      """
+
+  Scenario: Extract Sentences considers scenario description
+    Given a file named "foo.feature" with:
+      """
+      Feature: Foo
+        As a user,
+        I want something
+        so that I have that
+
+        Scenario: Bar
+        This is a sentence description
+        
+          Given a foo
+          When I bar
+          Then I baz
+      """
+    When I run `ruby extract_sentences.rb`
+    Then it should pass with:
+      """
+      Foo
+      As a user,
+      I want something
+      so that I have that
+      Bar
+      This is a sentence description
+      Given a foo when I bar then I baz
+      """

data/features/support/env.rb

@@ -0,0 +1,5 @@
+require 'aruba/cucumber'
+
+Before do
+  @aruba_timeout_seconds = 1000  # external java process, give it some time
+end

data/features/tag.feature

@@ -0,0 +1,53 @@
+Feature: Tag
+  As a Business Analyst
+  I want to tag all words from my features files
+  so that I can build up a glossary or check against forbidden words
+
+  Background: Prepare Testee
+    Given a file named "tag.rb" with:
+      """
+      $LOAD_PATH << '../../lib'
+      require 'gherkin_language'
+
+      no_cache = true
+      language = GherkinLanguage.new no_cache
+      puts language.tag ['tag.feature']
+
+      """
+
+  Scenario: Tag words
+    Given a file named "tag.feature" with:
+      """
+      Feature: Test
+        Scenario: Tag
+          Given a test
+          When execute
+          Then pass
+      """
+    When I run `ruby tag.rb`
+    Then it should pass with exactly:
+      """
+      <S>
+      [[[/null]
+      "["/``]
+      Test[Test/NNP]
+      "["/'']
+      ,[,/,,O]
+      "["/``,O]
+      Tag[tag/NN:UN,tag/VB,tag/VBP]
+      "["/'']
+      ,[,/,,O]
+      "["/``,O]
+      Given[Given/NNP,B-VP]
+      a[a/DT,B-NP-singular]
+      test[test/JJ,test/NN,test/VB,test/VBP,E-NP-singular]
+      when[when/WRB,B-ADVP]
+      execute[execute/VB,execute/VBP,B-VP]
+      then[then/JJ,then/NN,then/RB,I-VP]
+      pass[pass/JJ,pass/NN,pass/VB,pass/VBP,I-VP]
+      "["/'',B-ADVP]
+      ]
+      [</S>,O]
+
+
+      """

data/gherkin_language.gemspec

@@ -0,0 +1,14 @@
+Gem::Specification.new do |s|
+  s.name        = 'gherkin_language'
+  s.version     = '0.0.2'
+  s.date        = '2015-07-02'
+  s.summary     = 'Gherkin Language'
+  s.description = 'Check language of Gherkin Files'
+  s.authors     = ['Stefan Rohe']
+  s.homepage    = 'http://github.com/funkwerk/gherkin_language/'
+  s.files       = `git ls-files`.split("\n")
+  s.executables = s.files.grep(%r{^bin/}) { |file| File.basename(file) }
+  s.add_runtime_dependency 'gherkin', ['>= 2.12.2']
+  s.add_runtime_dependency 'term-ansicolor', ['>= 1.3.2']
+  s.add_development_dependency 'aruba', ['>= 0.6.2']
+end

data/lib/gherkin_language.rb

@@ -0,0 +1,333 @@
+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'
+
+# 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'],
+          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)
+    path = "~/.gherkin_language/#{LanguageToolProcess::VERSION}/accepted_paragraphs.yml"
+    @settings_path = File.expand_path path
+    @accepted_paragraphs = {}
+    begin
+      @accepted_paragraphs = YAML.load_file @settings_path unless no_cache
+    rescue
+      puts 'could not read settings'
+    end
+    @references = {}
+    @line_to_reference = {}
+  end
+
+  def analyze(file)
+    sentences = extract_sentences parse file
+    sentences.select! { |sentence| !accepted? sentence }
+    return if sentences.empty?
+    sentences.each do |sentence|
+      stripped = sentence.strip
+      @references[stripped] = [] unless @references.include? stripped
+      @references[stripped].push file
+    end
+  end
+
+  def accepted?(sentence)
+    return false if @accepted_paragraphs.nil?
+    key = :without_glossary
+    key = hash(File.read '.glossary') if File.exist? '.glossary'
+
+    return false unless @accepted_paragraphs.key? key
+    @accepted_paragraphs[key].include? hash sentence
+  end
+
+  def hash(value)
+    Digest::MD5.digest value.strip
+  end
+
+  def parse(file)
+    content = File.read file
+    to_json(content, file)
+  end
+
+  def extract_sentences(parsed)
+    feature_names = lambda do |input|
+      input.map { |feature| feature['name'] unless feature['name'] == '' }
+    end
+
+    descriptions = lambda do |input|
+      input.map { |feature| feature['description'] unless feature['description'] == '' }
+    end
+
+    sentences = feature_names.call(parsed) + descriptions.call(parsed) + scenario_names(parsed) + sentences(parsed)
+    sentences.select! { |sentence| sentence }
+    sentences.map { |sentence| sentence.gsub(/ «.+»/, '') }
+  end
+
+  def tag(files)
+    sentences = files.map { |file| extract_sentences parse file }
+    language = LanguageToolProcess.new
+    language.tag sentences
+  end
+
+  def report
+    return if @references.keys.empty?
+    language = LanguageToolProcess.new
+    language.start!
+
+    @references.keys.each do |sentence|
+      location = language.check_paragraph sentence
+      location.map { |line| @line_to_reference[line] = sentence }
+    end
+    language.stop!
+    errors = language.errors
+    unknown_words = language.unknown_words
+
+    used_refs = Set.new []
+    errors.each do |error|
+      used_refs.add @line_to_reference[error.from_y]
+      local_refs = @references[@line_to_reference[error.from_y]]
+      puts error.str local_refs
+    end
+    # TODO: list references for unknown words
+    puts red "#{unknown_words.count} unknown words: #{unknown_words * ', '}" unless unknown_words.empty?
+    return -1 unless unknown_words.empty?
+
+    @references.each do |sentence, _refs|
+      next if used_refs.include? sentence
+      key = :without_glossary
+      key = hash(File.read '.glossary') if File.exist? '.glossary'
+
+      @accepted_paragraphs[key] = Set.new [] unless @accepted_paragraphs.key? key
+      @accepted_paragraphs[key].add hash sentence
+    end
+
+    FileUtils.mkdir_p File.dirname @settings_path
+    File.open(@settings_path, 'w') do |settings_file|
+      settings_file.write @accepted_paragraphs.to_yaml
+    end
+    return -1 unless errors.empty?
+    0
+  end
+
+  def to_json(input, file = 'generated.feature')
+    io = StringIO.new
+    formatter = Gherkin::Formatter::JSONFormatter.new(io)
+    parser = Gherkin::Parser::Parser.new(formatter, true)
+    parser.parse(input, file, 0)
+    formatter.done
+    MultiJson.load io.string
+  end
+
+  def scenario_names(input)
+    # TODO: scenario outlines with example values inside?
+    scenarios = []
+    input.each do |features|
+      next unless features.key? 'elements'
+      elements = features['elements']
+      elements.each do |scenario|
+        scenarios.push scenario['name'] if scenario['type'] == 'scenario'
+        scenarios.push scenario['name'] if scenario['type'] == 'scenario_outline'
+        scenarios.push scenario['description'] unless scenario['description'].empty?
+      end
+    end
+    scenarios
+  end
+
+  def sentences(input)
+    sentences = []
+    background = []
+    input.each do |features|
+      next unless features.key? 'elements'
+      elements = features['elements']
+      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
+          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 ' '
+        if scenario.key? 'examples'
+          # TODO: support for multiple examples?
+          scenario['examples'].each do |example|
+            sentences.push example['name'] unless example['name'].empty?
+            sentences.push example['description'] unless example['description'].empty?
+            expand_outlines(sentence.strip, example).map { |expanded| sentences.push expanded }
+          end
+        else
+          sentences.push sentence.strip
+        end
+      end
+    end
+    sentences
+  end
+
+  def uncapitalize(term)
+    term[0, 1].downcase + term[1..-1]
+  end
+
+  def expand_outlines(sentence, example)
+    result = []
+    headers = example['rows'][0]['cells']
+    example['rows'].slice(1, example['rows'].length).each do |row|
+      modified_sentence = sentence.dup
+      headers.zip(row['cells']).map { |key, value| modified_sentence.gsub!("<#{key}>", value) }
+      result.push modified_sentence
+    end
+    result
+  end
+end

data/test/test_gherkin_language.rb

@@ -0,0 +1,61 @@
+require 'minitest/autorun'
+require 'gherkin_language'
+
+# gherkin language test
+class GherkinLanguageTest < Minitest::Unit::TestCase
+  def setup
+    @gherkin = GherkinLanguage.new
+  end
+
+  def test_to_json
+    # setup
+    feature = %(Feature: Foo
+Scenario: Bar
+)
+    expected = MultiJson.load %(
+      [
+       {
+        "keyword": "Feature",
+        "name": "Foo",
+        "line": 1,
+        "description": "",
+        "id": "foo",
+        "uri": "generated.feature",
+        "elements":
+         [
+          {
+           "keyword": "Scenario", "name": "Bar",
+           "line": 2, "description": "",
+           "id": "foo;bar", "type": "scenario"
+          }
+         ]
+        }
+       ]
+      )
+
+    # exercise
+    actual = @gherkin.to_json feature
+
+    # verify
+    assert_equal expected, actual
+  end
+
+  def test_sentences
+    # setup
+    feature = %(Feature: Foo
+Scenario: Bar
+  Given a Foo
+  And another Foo
+  But no Bar
+  When a Bar
+  Then a Baz
+)
+    expected = ['Given a Foo and another Foo but no Bar when a Bar then a Baz']
+
+    # exercise
+    actual = @gherkin.sentences @gherkin.to_json feature
+
+    # verify
+    assert_equal expected, actual
+  end
+end

data/test/test_gherkin_language_tool.rb

@@ -0,0 +1,40 @@
+require 'minitest/autorun'
+require 'gherkin_language'
+
+# checks language
+class LanguageTest < Minitest::Unit::TestCase
+  def setup
+    @language = GherkinLanguage::LanguageToolProcess.new
+    @language.start!
+  end
+
+  def test_check_valid_paragraph
+    # setup
+    feature = 'This is good English'
+
+    # exercise
+    @language.check_paragraph feature
+    @language.stop!
+    actual = @language.errors
+    unknown_words = @language.unknown_words
+
+    # verify
+    assert_empty actual
+    assert_empty unknown_words
+  end
+
+  def test_check_invalid_paragraph
+    # setup
+    feature = 'This are no English good'
+
+    # exercise
+    @language.check_paragraph feature
+    @language.stop!
+    actual = @language.errors
+    unknown_words = @language.unknown_words
+
+    # verify
+    refute_empty actual
+    assert_empty unknown_words
+  end
+end

metadata

@@ -0,0 +1,101 @@
+--- !ruby/object:Gem::Specification
+name: gherkin_language
+version: !ruby/object:Gem::Version
+  version: 0.0.2
+platform: ruby
+authors:
+- Stefan Rohe
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2015-07-02 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: gherkin
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 2.12.2
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 2.12.2
+- !ruby/object:Gem::Dependency
+  name: term-ansicolor
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.3.2
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 1.3.2
+- !ruby/object:Gem::Dependency
+  name: aruba
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.6.2
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: 0.6.2
+description: Check language of Gherkin Files
+email: 
+executables:
+- gherkin_language
+extensions: []
+extra_rdoc_files: []
+files:
+- ".rubocop.yml"
+- ".travis.yml"
+- Gemfile
+- Guardfile
+- LICENSE
+- README.md
+- Rakefile
+- bin/gherkin_language
+- features/report.feature
+- features/sentences.feature
+- features/support/env.rb
+- features/tag.feature
+- gherkin_language.gemspec
+- lib/gherkin_language.rb
+- test/test_gherkin_language.rb
+- test/test_gherkin_language_tool.rb
+homepage: http://github.com/funkwerk/gherkin_language/
+licenses: []
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 2.2.3
+signing_key: 
+specification_version: 4
+summary: Gherkin Language
+test_files: []