Class: GherkinLanguage

Inherits:
Object
  • Object
show all
Defined in:
lib/gherkin_language.rb

Overview

gherkin utilities

Instance Method Summary collapse

Constructor Details

#initialize(no_cache = false, ngram = false, unknown_words = false) ⇒ GherkinLanguage

Returns a new instance of GherkinLanguage.



21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
 
# File 'lib/gherkin_language.rb', line 21

def initialize(no_cache = false, ngram = false, unknown_words = 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 = {}
  @exceptions = %w(SENTENCE_FRAGMENT ENGLISH_WORD_REPEAT_BEGINNING_RULE)
  @ngram = ngram
  @unknown_words = unknown_words
end

Instance Method Details

#accepted?(sentence) ⇒ Boolean

Returns:

  • (Boolean) 


79
80
81
82
83
84
85
86
 
# File 'lib/gherkin_language.rb', line 79

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

#analyze(file) ⇒ Object 



41
42
43
44
45
46
47
48
49
50
 
# File 'lib/gherkin_language.rb', line 41

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

#determine_readability_by_file(files) ⇒ Object 



52
53
54
55
56
57
58
59
60
61
62
63
64
65
 
# File 'lib/gherkin_language.rb', line 52

def determine_readability_by_file(files)
  puts "Readability. Sorted from best to worst readable feature\n\n" if files.length > 1
  readability_by_file = {}
  files.each do |file|
    sentences = extract_sentences parse(file)
    readability_by_file[file] = readability sentences
  end
  average_readability = 0
  readability_by_file.sort { |lhs, rhs| lhs[1] <=> rhs[1] }.reverse_each do |file, rating|
    puts "#{rating.round}: #{file}"
    average_readability += rating / files.length
  end
  puts "\n#{files.length} files analyzed. Average readability is #{average_readability.round}" if files.length > 1
end

#expand_outlines(sentence, example) ⇒ Object 



241
242
243
244
245
246
247
248
249
250
 
# File 'lib/gherkin_language.rb', line 241

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

#extract_examples(examples, prototype) ⇒ Object 



227
228
229
230
231
232
233
234
235
 
# File 'lib/gherkin_language.rb', line 227

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

#extract_sentences(parsed) ⇒ Object 



97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
 
# File 'lib/gherkin_language.rb', line 97

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(/ an «[^«]*»/, ' a replacement') }
  sentences.map! { |sentence| sentence.gsub(/ ?«[^«]*»/, ' replacement') }
  sentences.map { |sentence| sentence.gsub('replacement replacement', 'replacement') }
end

#extract_terms_from_scenario(steps, background) ⇒ Object 



216
217
218
219
220
221
222
223
224
225
 
# File 'lib/gherkin_language.rb', line 216

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

#hash(value) ⇒ Object 



88
89
90
 
# File 'lib/gherkin_language.rb', line 88

def hash(value)
  Digest::MD5.digest value.strip
end

#ignore(exception) ⇒ Object 



37
38
39
 
# File 'lib/gherkin_language.rb', line 37

def ignore(exception)
  @exceptions.push exception
end

#parse(file) ⇒ Object 



92
93
94
95
 
# File 'lib/gherkin_language.rb', line 92

def parse(file)
  content = File.read file
  to_json(content, file)
end

#readability(sentences) ⇒ Object 



67
68
69
70
71
72
73
74
75
76
77
 
# File 'lib/gherkin_language.rb', line 67

def readability(sentences)
  require 'syllables'

  total_words = 0
  total_syllabels = 0
  Syllables.new(sentences.join('\n')).to_h.each do |_word, syllabels|
    total_words += 1
    total_syllabels += syllabels
  end
  206.835 - 1.015 * (total_words / sentences.length) - 84.6 * (total_syllabels / total_words)
end

#report ⇒ Object 



119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
 
# File 'lib/gherkin_language.rb', line 119

def report
  return 0 if @references.keys.empty?
  language = LanguageToolProcess.new(@ngram, @unknown_words)
  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]
  end
  errors.select! { |error| !@exceptions.include? error.rule }
  errors.each do |error|
    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?

  write_accepted_paragraphs used_refs

  return -1 unless errors.empty?
  0
end

#scenario_names(input) ⇒ Object 



176
177
178
179
180
181
182
183
184
185
186
187
188
189
 
# File 'lib/gherkin_language.rb', line 176

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

#sentences(input) ⇒ Object 



191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
 
# File 'lib/gherkin_language.rb', line 191

def sentences(input)
  sentences = []
  background = []
  input.each do |features|
    next unless features.key? 'elements'
    features['elements'].each do |scenario|
      next unless scenario.key? 'steps'
      terms = background.dup
      if scenario['type'] == 'background'
        background.push extract_terms_from_scenario(scenario['steps'], terms)
        next
      end

      terms.push extract_terms_from_scenario(scenario['steps'], background)
      sentence = terms.join(' ').strip
      if scenario.key? 'examples'
        sentences += extract_examples(scenario['examples'], sentence)
      else
        sentences.push sentence
      end
    end
  end
  sentences
end

#tag(files) ⇒ Object 



113
114
115
116
117
 
# File 'lib/gherkin_language.rb', line 113

def tag(files)
  sentences = files.map { |file| extract_sentences parse file }
  language = LanguageToolProcess.new
  language.tag sentences
end

#to_json(input, file = 'generated.feature') ⇒ Object 



167
168
169
170
171
172
173
174
 
# File 'lib/gherkin_language.rb', line 167

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

#uncapitalize(term) ⇒ Object 



237
238
239
 
# File 'lib/gherkin_language.rb', line 237

def uncapitalize(term)
  term[0, 1].downcase + term[1..-1]
end

#write_accepted_paragraphs(used_refs) ⇒ Object 



151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
 
# File 'lib/gherkin_language.rb', line 151

def write_accepted_paragraphs(used_refs)
  @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
end