gherkin 0.0.3

data/ragel/lexer.java.rl.erb

@@ -0,0 +1,198 @@
+package gherkin.lexer;
+
+import java.util.List;
+import java.util.ArrayList;
+import java.util.regex.Pattern;
+import gherkin.Lexer;
+import gherkin.Listener;
+import gherkin.LexingError;
+
+public class <%= i18n_lexer_class_name %> implements Lexer {
+  %%{
+    machine lexer;
+    alphtype byte;
+
+    action begin_content {
+      contentStart = p;
+      currentLine = lineNumber;
+    }
+
+    action start_pystring {
+      currentLine = lineNumber;
+      startCol = p - lastNewline;
+    }
+    
+    action begin_pystring_content {
+      contentStart = p;
+    }
+
+    action store_pystring_content {
+      String con = unindent(startCol, substring(data, contentStart, nextKeywordStart-1).replaceFirst("(\\r?\\n)?( )*\\Z", ""));
+      listener.py_string(con, currentLine);
+    }
+
+    action store_feature_content {
+      String con = multilineStrip(keywordContent(data, p, eof, nextKeywordStart, contentStart).trim());
+      listener.feature(keyword, con, currentLine);
+      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
+      nextKeywordStart = -1;
+    }
+
+    action store_background_content {
+      String con = multilineStrip(keywordContent(data, p, eof, nextKeywordStart, contentStart));
+      listener.background(keyword, con, currentLine);
+      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
+      nextKeywordStart = -1;
+    }
+
+    action store_scenario_content {
+      String con = multilineStrip(keywordContent(data, p, eof, nextKeywordStart, contentStart));
+      listener.scenario(keyword, con, currentLine);
+      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
+      nextKeywordStart = -1;
+    }
+
+    action store_scenario_outline_content {
+      String con = multilineStrip(keywordContent(data, p, eof, nextKeywordStart, contentStart));
+      listener.scenario_outline(keyword, con, currentLine);
+      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
+      nextKeywordStart = -1;
+    }
+
+    action store_examples_content {
+      String con = multilineStrip(keywordContent(data, p, eof, nextKeywordStart, contentStart));
+      listener.examples(keyword, con, currentLine);
+      if(nextKeywordStart != -1) p = nextKeywordStart - 1;
+      nextKeywordStart = -1;
+    }
+
+    action store_step_content {
+      listener.step(keyword, substring(data, contentStart, p).trim(), currentLine);
+    }
+
+    action store_comment_content {
+      listener.comment(substring(data, contentStart, p).trim(), lineNumber);
+    }
+
+    action store_tag_content {
+      listener.tag(substring(data, contentStart, p).trim(), currentLine);
+    }
+
+    action inc_line_number {
+      lineNumber++;
+    }
+
+    action last_newline {
+      lastNewline = p + 1;
+    }
+
+    action start_keyword {
+      if(keywordStart == -1) keywordStart = p;
+    }
+
+    action end_keyword {
+      keyword = substring(data, keywordStart, p).replaceFirst(":$","").trim();
+      keywordStart = -1;
+    }
+
+    action next_keyword_start {
+      nextKeywordStart = p;
+    }
+
+    action start_table {
+      p = p - 1;
+      rows = new ArrayList<List<String>>();
+      currentLine = lineNumber;
+    }
+
+    action start_row {
+      currentRow = new ArrayList<String>();
+    }
+
+    action begin_cell_content {
+      contentStart = p;
+    }
+
+    action store_cell_content {
+      currentRow.add(substring(data, contentStart, p).trim());
+    }
+
+    action store_row {
+      rows.add(currentRow);
+    }
+
+    action store_table {
+      if(!rows.isEmpty()) {
+        listener.table(rows, currentLine);
+      }
+    }
+
+    action end_feature {
+      if(cs < lexer_first_final) {
+        String content = currentLineContent(data, lastNewline);
+        throw new LexingError("Lexing error on line " + lineNumber);
+      }
+    }
+
+    include lexer_common "lexer_common.<%= i18n_language %>.rl"; 
+  }%%
+
+    private final Listener listener;
+
+    public <%= i18n_lexer_class_name %>(Listener listener) {
+        this.listener = listener;
+    }
+
+    %% write data noerror;
+
+    public void scan(CharSequence inputSequence) {
+        String input = inputSequence.toString() + "\n%_FEATURE_END_%";
+        byte[] data = input.getBytes();
+        int cs, p = 0, pe = data.length;
+        int eof = pe;
+
+        int lineNumber = 1;
+        int lastNewline = 0;
+
+        int contentStart = -1;
+        int currentLine = -1;
+        int startCol = -1;
+        int nextKeywordStart = -1;
+        int keywordStart = -1;
+        String keyword = null;
+        List<List<String>> rows = null;
+        List<String> currentRow = null;
+
+        %% write init;
+        %% write exec;
+    }
+
+    private String keywordContent(byte[] data, int p, int eof, int nextKeywordStart, int contentStart) {
+      int endPoint = (nextKeywordStart == -1 || (p == eof)) ? p : nextKeywordStart;
+      return substring(data, contentStart, endPoint);
+    }
+
+    private String multilineStrip(String text) {
+        StringBuffer result = new StringBuffer();
+        for(String s : text.split("\n")) {
+            result.append(s.trim()).append("\n");
+        }
+        return result.toString().trim();
+    }
+
+    private String unindent(int startCol, String text) {
+        return Pattern.compile("^ {0," + startCol + "}", Pattern.MULTILINE).matcher(text).replaceAll("");
+    }
+
+    private String currentLineContent(byte[] data, int lastNewline) {
+        return substring(data, lastNewline, data.length).trim();
+    }
+
+    private String substring(byte[] data, int start, int end) {
+        try {
+            return new String(data, start, end-start, "utf-8");
+        } catch(java.io.UnsupportedEncodingException e) {
+            throw new RuntimeException("Internal error", e);
+        }
+    }
+}

data/ragel/lexer.rb.rl.erb

@@ -0,0 +1,172 @@
+module Gherkin
+  module RbLexer
+    class <%= i18n_lexer_class_name %> #:nodoc:
+      %%{
+        # patterns:
+        # * data[start...end].pack("c*").strip_of_some_sort
+        # * changing the end point of the range according to next_keyword_start
+        # * methods taking the machine state because Ragel doesn't seem to know about ivars
+ 
+        machine lexer;
+ 
+        action begin_content {
+          @content_start = p
+          @current_line = @line_number
+        }
+
+        action start_pystring {
+          @current_line = @line_number
+          @start_col = p - @last_newline
+        }
+        
+        action begin_pystring_content {
+          @content_start = p
+        }
+
+        action store_pystring_content {
+          con = unindent(@start_col, data[@content_start...@next_keyword_start-1].pack("c*").sub(/(\r?\n)?( )*\Z/, ''))
+          @listener.py_string(con, @current_line) 
+        }
+ 
+        action store_feature_content {
+          store_keyword_content(:feature, data, p, eof) { |con| multiline_strip(con) }
+          p = @next_keyword_start - 1 if @next_keyword_start          
+          @next_keyword_start = nil
+        }
+ 
+        action store_background_content {
+          store_keyword_content(:background, data, p, eof) { |con| multiline_strip(con) }
+          p = @next_keyword_start - 1 if @next_keyword_start                    
+          @next_keyword_start = nil
+        }
+
+        action store_scenario_content {
+          store_keyword_content(:scenario, data, p, eof) { |con| multiline_strip(con) }
+          p = @next_keyword_start - 1 if @next_keyword_start                    
+          @next_keyword_start = nil
+        }
+
+        action store_scenario_outline_content {
+          store_keyword_content(:scenario_outline, data, p, eof) { |con| multiline_strip(con) }
+          p = @next_keyword_start - 1 if @next_keyword_start                    
+          @next_keyword_start = nil
+        }
+
+        action store_examples_content {
+          store_keyword_content(:examples, data, p, eof) { |con| multiline_strip(con) }
+          p = @next_keyword_start - 1 if @next_keyword_start                    
+          @next_keyword_start = nil
+        }
+
+        action store_step_content {
+          con = data[@content_start...p].pack("c*").strip
+          @listener.step(@keyword, con, @current_line)
+        }
+ 
+        action store_comment_content {
+          con = data[@content_start...p].pack("c*").strip
+          @listener.comment(con, @line_number)
+        }
+ 
+        action store_tag_content {
+          con = data[@content_start...p].pack("c*").strip
+          @listener.tag(con, @current_line)
+        }
+ 
+        action inc_line_number {
+          @line_number += 1
+        }
+ 
+        action last_newline {
+          @last_newline = p + 1
+        }
+
+        action start_keyword {
+          @keyword_start ||= p
+        }
+ 
+        action end_keyword {
+          @keyword = data[@keyword_start...p].pack("c*").sub(/:$/,'').strip
+          @keyword_start = nil
+        }
+ 
+        action next_keyword_start {
+          @next_keyword_start = p
+        }
+
+        action start_table {
+          p = p - 1
+          @rows = []
+          @current_line = @line_number
+        }
+
+        action start_row {
+          current_row = []
+        }
+ 
+        action begin_cell_content {
+          @content_start = p
+        }
+
+        action store_cell_content {
+          con = data[@content_start...p].pack("c*").strip
+          current_row << con
+        }
+
+        action store_row {
+          @rows << current_row
+        }
+
+        action store_table {
+          if @rows.size != 0
+            @listener.table(@rows, @current_line)
+          end 
+        }
+ 
+        action end_feature {
+          if cs < lexer_first_final
+            content = current_line_content(data, p)
+            raise Lexer::LexingError.new("Lexing error on line %d: '%s'." % [@line_number, content])
+          end
+        }
+
+        include lexer_common "lexer_common.<%= i18n_language %>.rl"; 
+      }%%
+ 
+      def initialize(listener)
+        @listener = listener
+        %% write data;
+      end
+ 
+      def scan(data)
+        data = (data + "\n%_FEATURE_END_%").unpack("c*") # Explicit EOF simplifies things considerably
+        eof = pe = data.length
+ 
+        @line_number = 1
+        @last_newline = 0
+ 
+        %% write init;
+        %% write exec;
+      end
+ 
+      def multiline_strip(text)
+        text.split("\n").map{|s| s.strip}.join("\n").strip
+      end
+
+      def unindent(startcol, text)
+        text.gsub(/^ {0,#{startcol}}/, "")
+      end
+ 
+      def store_keyword_content(event, data, p, eof)
+        end_point = (!@next_keyword_start or (p == eof)) ? p : @next_keyword_start
+        con = yield data[@content_start...end_point].pack("c*")
+        @listener.send(event, @keyword, con, @current_line)
+      end
+      
+      def current_line_content(data, p)
+        rest = data[@last_newline..-1]
+        rest[0..rest.index(10)||-1].pack("c*").strip
+      end
+    end
+  end
+end

data/ragel/lexer_common.rl.erb

@@ -0,0 +1,46 @@
+%%{
+  machine lexer_common;
+
+  # Language specific
+  I18N_Feature = <%= i18n['feature'] %> >start_keyword %end_keyword; 
+  I18N_Background = <%= i18n['background'] %> >start_keyword %end_keyword;
+  I18N_ScenarioOutline = <%= i18n['scenario_outline'] %> >start_keyword %end_keyword;
+  I18N_Scenario = <%= i18n['scenario'] %> >start_keyword %end_keyword;
+  I18N_Step = (<%= i18n['given'] %> | <%= i18n['when'] %> | <%= i18n['and'] %> | <%= i18n['then'] %> | <%= i18n['but'] %>) >start_keyword %end_keyword;
+  I18N_Examples = <%= i18n['examples'] %> >start_keyword %end_keyword;
+
+  EOF = '%_FEATURE_END_%'; # Explicit EOF added before scanning begins 
+  EOL = ('\r'? '\n') @inc_line_number @last_newline;
+
+  FeatureHeadingEnd = EOL+ space* (I18N_Background | I18N_Scenario | I18N_ScenarioOutline | '@' | '#' | EOF) >next_keyword_start;
+  ScenarioHeadingEnd = EOL+ space* ( I18N_Scenario | I18N_ScenarioOutline | I18N_Step | '@' | '#' | EOF ) >next_keyword_start;
+  BackgroundHeadingEnd = EOL+ space* ( I18N_Scenario | I18N_ScenarioOutline | I18N_Step | '@' | '#'| EOF ) >next_keyword_start;
+  ScenarioOutlineHeadingEnd = EOL+ space* ( I18N_Scenario | I18N_Step | '@' | '#' | EOF ) >next_keyword_start;
+  ExamplesHeadingEnd = EOL+ space* '|' >next_keyword_start;
+ 
+  FeatureHeading = space* I18N_Feature %begin_content ^FeatureHeadingEnd* :>> FeatureHeadingEnd @store_feature_content;
+  BackgroundHeading = space* I18N_Background %begin_content ^BackgroundHeadingEnd* :>> BackgroundHeadingEnd @store_background_content;
+  ScenarioHeading = space* I18N_Scenario %begin_content ^ScenarioHeadingEnd* :>> ScenarioHeadingEnd @store_scenario_content;
+  ScenarioOutlineHeading = space* I18N_ScenarioOutline %begin_content ^ScenarioOutlineHeadingEnd* :>> ScenarioOutlineHeadingEnd @store_scenario_outline_content;
+  ExamplesHeading = space* I18N_Examples %begin_content ^ExamplesHeadingEnd* :>> ExamplesHeadingEnd @store_examples_content;
+
+  Step = space* I18N_Step %begin_content ^EOL+ %store_step_content EOL+;
+  Comment = space* '#' >begin_content ^EOL* %store_comment_content EOL+;
+  
+  Tag = ( '@' [^@\r\n\t ]+ >begin_content ) %store_tag_content;
+  Tags = space* (Tag space*)+ EOL+;
+  
+  StartTable = space* '|' >start_table;
+  EndTable = EOL space* ^('|') >next_keyword_start;    
+  Cell = '|' (any - '|')* >begin_cell_content %store_cell_content;
+  Row = space* Cell* >start_row '|' :>> (space* EOL+ space*) %store_row;
+  Table = StartTable :>> Row+ %store_table <: EndTable?;
+  
+  StartPyString = '"""' >start_pystring space* :>> EOL; 
+  EndPyString = (space* '"""') >next_keyword_start; 
+  PyString = space* StartPyString %begin_pystring_content (^EOL | EOL)* :>> EndPyString %store_pystring_content space* EOL+;
+  
+  Tokens = (space | EOL)* (Tags | Comment | FeatureHeading | BackgroundHeading | ScenarioHeading | ScenarioOutlineHeading | ExamplesHeading | Step | Table | PyString)* (space | EOL)* EOF;
+
+  main := Tokens %end_feature @!end_feature;
+}%%

data/spec/gherkin/c_lexer_spec.rb

@@ -0,0 +1,21 @@
+#encoding: utf-8
+unless defined?(JRUBY_VERSION)
+require File.expand_path(File.dirname(__FILE__) + '/../spec_helper')
+require 'gherkin/c_lexer'
+
+module Gherkin
+  module Lexer
+    describe "C Lexer" do
+      before do
+        @listener = Gherkin::SexpRecorder.new
+        @lexer = Gherkin::CLexer['en'].new(@listener)
+      end
+
+      it_should_behave_like "a Gherkin lexer"
+      it_should_behave_like "a Gherkin lexer lexing tags"
+      it_should_behave_like "a Gherkin lexer lexing py_strings"
+      it_should_behave_like "a Gherkin lexer lexing tables"
+    end
+  end
+end
+end

data/spec/gherkin/fixtures/complex.feature

@@ -0,0 +1,43 @@
+#Comment on line 1
+@tag1 @tag2
+#Comment on line 3
+Feature: Feature Text
+  In order to test multiline forms
+  As a ragel writer
+  I need to check for complex combinations
+
+  #Comment on line 9
+
+  #Comment on line 11
+
+  Background:
+    Given this is a background step
+    And this is another one
+
+  @tag3 @tag4
+  Scenario: Reading a Scenario
+    Given there is a step
+    But not another step
+
+  @tag3
+  Scenario: Reading a second scenario
+    #Comment on line 24
+    Given a third step with a table
+    |a|b|
+    |c|d|
+    |e|f|
+    And I am still testing things
+      |g|h|
+      |e|r|
+      |k|i|
+      |n|| 
+    And I am done testing these tables
+    #Comment on line 29
+    Then I am happy
+
+  Scenario: Hammerzeit
+    Given All work and no play
+      """
+      Makes Homer something something
+      """
+    Then crazy