gherkin 1.0.30-universal-dotnet

data/lib/gherkin/parser/root.txt

@@ -0,0 +1,11 @@
+ |                  | feature | background | scenario | scenario_outline | examples | step         | row            | py_string    | eof | comment    | tag        |
+ | root             | feature | E          | E        | E                | E        | E            | E              | E            | eof | push(meta) | push(meta) |
+ | feature          | E       | background | scenario | scenario_outline | E        | E            | E              | E            | eof | push(meta) | push(meta) |
+ | step             | E       | E          | scenario | scenario_outline | E        | step         | step           | step         | eof | push(meta) | push(meta) |
+ | outline_step     | E       | E          | scenario | scenario_outline | examples | outline_step | outline_step   | outline_step | eof | push(meta) | push(meta) |
+ | background       | E       | E          | scenario | scenario_outline | E        | step         | E              | E            | eof | push(meta) | push(meta) |
+ | scenario         | E       | E          | scenario | scenario_outline | E        | step         | E              | E            | eof | push(meta) | push(meta) |
+ | scenario_outline | E       | E          | E        | E                | E        | outline_step | E              | E            | eof | push(meta) | push(meta) |
+ | examples         | E       | E          | E        | E                | E        | E            | examples_table | E            | eof | push(meta) | push(meta) |
+ | examples_table   | E       | E          | scenario | scenario_outline | examples | E            | examples_table | E            | eof | push(meta) | push(meta) |
+ | eof              | E       | E          | E        | E                | E        | E            | E              | E            | E   | E          | E          |

data/lib/gherkin/parser/steps.txt

@@ -0,0 +1,4 @@
+ |       | feature | background | scenario | scenario_outline | examples | step | row  | py_string | eof | comment | tag |
+ | steps | E       | E          | E        | E                | E        | step | E    | E         | eof | E       | E   |
+ | step  | E       | E          | E        | E                | E        | step | step | steps     | eof | E       | E   |
+ | eof   | E       | E          | E        | E                | E        | E    | E    | E         | E   | E       | E   |

data/lib/gherkin/parser/tag_expression.rb

@@ -0,0 +1,50 @@
+require 'gherkin/native'
+
+module Gherkin
+  module Parser
+    class TagExpression
+      native_impl('gherkin')
+
+      attr_reader :limits
+
+      def initialize(tag_expressions)
+        @ands = []
+        @limits = {}
+        tag_expressions.each do |expr|
+          add(expr.strip.split(/\s*,\s*/))
+        end
+      end
+
+      def empty?
+        @ands.empty?
+      end
+
+      def eval(tags)
+        return true if @ands.flatten.empty?
+        vars = Hash[*tags.map{|tag| [tag, true]}.flatten]
+        !!Kernel.eval(ruby_expression)
+      end
+
+    private
+
+      def add(tags)
+        negatives, positives = tags.partition{|tag| tag =~ /^~/}
+        positive_limits = Hash[*positives.map{|positive| tag, limit = positive.split(':'); [tag, limit ? limit.to_i : nil]}.flatten]
+        @limits.merge!(positive_limits)
+        @ands << (negatives + positive_limits.keys)
+      end
+
+      def ruby_expression
+        "(" + @ands.map do |ors|
+          ors.map do |tag|
+            if tag =~ /^~(.*)/
+              "!vars['#{$1}']"
+            else
+              "vars['#{tag}']"
+            end
+          end.join("||")
+        end.join(")&&(") + ")"
+      end
+    end
+  end
+end

data/lib/gherkin/rb_lexer.rb

@@ -0,0 +1,8 @@
+module Gherkin
+  module RbLexer
+    def self.[](i18n_underscored_iso_code)
+      require "gherkin/rb_lexer/#{i18n_underscored_iso_code}"
+      const_get(i18n_underscored_iso_code.capitalize)
+    end
+  end
+end

data/lib/gherkin/rb_lexer/.gitignore

@@ -0,0 +1 @@
+*.rb

data/lib/gherkin/rb_lexer/README.rdoc

@@ -0,0 +1,8 @@
+= Lexers
+
+Gherkin support lexing of lots of natural languages, defined by gherkin/i18n.yml
+The lexers are generated with the following command:
+
+    rake ragel:i18n
+
+You have to run this command if you modify gherkin/i18n.yml

data/lib/gherkin/rubify.rb

@@ -0,0 +1,18 @@
+module Gherkin
+  module Rubify
+    if defined?(JRUBY_VERSION)
+      # Translate Java objects to Ruby.
+      def rubify(o)
+        if Java.java.util.Collection === o || Array === o
+          o.map{|e| rubify(e)}
+        else
+          o
+        end
+      end
+    else
+      def rubify(o)
+        o
+      end
+    end
+  end
+end

data/lib/gherkin/tools.rb

@@ -0,0 +1,8 @@
+module Gherkin
+  module Tools
+    SUB_COMMANDS = %w(stats reformat)
+    SUB_COMMANDS.each do |cmd|
+      autoload cmd.capitalize.to_sym, "gherkin/tools/#{cmd}"
+    end
+  end
+end

data/lib/gherkin/tools/files.rb

@@ -0,0 +1,35 @@
+require 'gherkin'
+
+module Gherkin
+  module Tools
+    # Base class for file based operations
+    class Files
+      include Enumerable
+
+      def initialize(paths)
+        raise "Please specify one or more paths" if paths.empty?
+        @paths = paths
+      end
+
+      def each(&proc)
+        globs = @paths.map do |path|
+          raise "#{path} does not exist" unless File.exist?(path)
+          File.directory?(path) ? File.join(path, '**', '*.feature') : path
+        end
+
+        Dir[*globs].uniq.sort.each(&proc)
+      end
+
+      def scan(file, listener)
+        parser = Gherkin::Parser::Parser.new(listener, true, "root")
+        lexer = Gherkin::I18nLexer.new(parser, false)
+        begin
+          lexer.scan(IO.read(file))
+        rescue => e
+          e.message << " (#{file})"
+          raise e
+        end
+      end
+    end
+  end
+end

data/lib/gherkin/tools/reformat.rb

@@ -0,0 +1,19 @@
+require 'stringio'
+require 'gherkin/tools/files'
+require 'gherkin/formatter/pretty_formatter'
+
+module Gherkin
+  module Tools
+    class Reformat < Files
+      def run
+        each do |file|
+          purdy = StringIO.new
+          listener = Formatter::PrettyFormatter.new(purdy)
+          scan(file, listener)
+          purdy.rewind
+          File.open(file, 'w') {|io| io.write(purdy.read)}
+        end
+      end
+    end
+  end
+end

data/lib/gherkin/tools/stats.rb

@@ -0,0 +1,21 @@
+require 'gherkin'
+require 'gherkin/tools/files'
+require 'gherkin/tools/stats_listener'
+
+module Gherkin
+  module Tools
+    class Stats < Files
+      def run
+        listener = StatsListener.new
+        each do |f|
+          parser = Gherkin::Parser::Parser.new(listener, true)
+          lexer = Gherkin::I18nLexer.new(parser)
+          lexer.scan(IO.read(f))
+        end
+        puts "Features:  #{listener.features}"
+        puts "Scenarios: #{listener.scenarios}"
+        puts "Steps:     #{listener.steps}"
+      end
+    end
+  end
+end

data/lib/gherkin/tools/stats_listener.rb

@@ -0,0 +1,57 @@
+require 'gherkin/native'
+
+# encoding: utf-8
+module Gherkin
+  module Tools
+    class StatsListener
+      implements 'gherkin.Listener'
+
+      attr_reader :features, :scenarios, :steps
+
+      def initialize
+        @features  = 0
+        @scenarios = 0
+        @steps     = 0
+      end
+
+      def tag(name, line)
+      end
+
+      def comment(content, line)
+      end
+
+      def feature(keyword, name, line)
+        @features += 1
+      end
+
+      def background(keyword, name, line)
+      end
+
+      def scenario(keyword, name, line)
+        @scenarios += 1
+      end
+
+      def scenario_outline(keyword, name, line)
+      end
+
+      def examples(keyword, name, line)
+      end
+
+      def step(keyword, name, line)
+        @steps += 1
+      end
+
+      def row(row, line)
+      end
+
+      def py_string(string, line)
+      end
+
+      def syntax_error(state, event, legal_events, line)
+      end
+
+      def eof
+      end
+    end
+  end
+end

data/ragel/i18n/.gitignore

@@ -0,0 +1 @@
+*.rl

data/ragel/lexer.c.rl.erb

@@ -0,0 +1,425 @@
+#include <assert.h>
+#include <ruby.h>
+
+#if defined(_WIN32)
+#include <stddef.h>
+#endif
+
+#ifdef HAVE_RUBY_RE_H
+#include <ruby/re.h>
+#else
+#include <re.h>
+#endif
+
+#ifdef HAVE_RUBY_ENCODING_H
+#include <ruby/encoding.h>
+#define ENCODED_STR_NEW(ptr, len) \
+    rb_enc_str_new(ptr, len, rb_utf8_encoding())
+#else
+#define ENCODED_STR_NEW(ptr, len) \
+    rb_str_new(ptr, len)
+#endif
+
+#define LF_FLAG      0
+#define CRLF_FLAG    1
+#define LF           "\n"
+#define CRLF         "\r\n"
+
+#ifndef RSTRING_PTR
+#define RSTRING_PTR(s) (RSTRING(s)->ptr)
+#endif
+
+#ifndef RSTRING_LEN
+#define RSTRING_LEN(s) (RSTRING(s)->len)
+#endif
+
+#define DATA_GET(FROM, TYPE, NAME) \
+  Data_Get_Struct(FROM, TYPE, NAME); \
+  if (NAME == NULL) { \
+    rb_raise(rb_eArgError, "NULL found for " # NAME " when it shouldn't be."); \
+  }
+ 
+typedef struct lexer_state {
+  int content_len;
+  int line_number;
+  int current_line;
+  int start_col;
+  int eol;
+  size_t mark;
+  size_t keyword_start;
+  size_t keyword_end;
+  size_t next_keyword_start;
+  size_t content_start;
+  size_t content_end;
+  size_t query_start;
+  size_t last_newline;
+  size_t final_newline;
+} lexer_state;
+
+static VALUE mGherkin;
+static VALUE mCLexer;
+static VALUE cI18nLexer;
+static VALUE rb_eGherkinLexingError;
+
+#define LEN(AT, P) (P - data - lexer->AT)
+#define MARK(M, P) (lexer->M = (P) - data)
+#define PTR_TO(P) (data + lexer->P)
+
+#define STORE_KW_END_CON(EVENT) \
+  store_kw_con(listener, # EVENT, \
+    PTR_TO(keyword_start), LEN(keyword_start, PTR_TO(keyword_end - 1)), \
+    PTR_TO(content_start), LEN(content_start, PTR_TO(content_end)), \
+    lexer->current_line, lexer->eol); \
+    if (lexer->content_end != 0) { \
+      p = PTR_TO(content_end - 1); \
+    } \
+    lexer->content_end = 0
+
+#define STORE_ATTR(ATTR) \
+    store_attr(listener, # ATTR, \
+      PTR_TO(content_start), LEN(content_start, p), \
+      lexer->line_number)
+
+%%{
+  machine lexer;
+ 
+  action begin_content {
+		MARK(content_start, p);
+    lexer->current_line = lexer->line_number;
+  }
+  
+  action begin_pystring_content {
+    MARK(content_start, p);
+  }
+
+  action start_pystring {
+    lexer->current_line = lexer->line_number;
+    lexer->start_col = p - data - lexer->last_newline;
+  }
+ 
+  action store_pystring_content {
+    int len = LEN(content_start, PTR_TO(final_newline));
+
+    if (len < 0) len = 0;
+
+    store_pystring_content(listener, lexer->start_col, PTR_TO(content_start), len, lexer->current_line);
+  } 
+ 
+  action store_feature_content {
+    STORE_KW_END_CON(feature);
+  }
+  
+  action store_background_content {
+    STORE_KW_END_CON(background);
+  }
+  
+  action store_scenario_content {
+    STORE_KW_END_CON(scenario);
+  }
+  
+  action store_scenario_outline_content {
+    STORE_KW_END_CON(scenario_outline);
+  }
+  
+  action store_examples_content {
+    STORE_KW_END_CON(examples);
+  }
+  
+  action store_step_content {
+    store_kw_con(listener, "step",
+      PTR_TO(keyword_start), LEN(keyword_start, PTR_TO(keyword_end)),
+      PTR_TO(content_start), LEN(content_start, p), 
+      lexer->current_line, lexer->eol);
+  }
+  
+  action store_comment_content {
+    STORE_ATTR(comment);
+    lexer->mark = 0;
+  }
+  
+  action store_tag_content {
+    STORE_ATTR(tag);
+    lexer->mark = 0;
+  }
+  
+  action inc_line_number {
+    lexer->line_number += 1;
+    MARK(final_newline, p);
+  }
+    
+  action last_newline {
+    MARK(last_newline, p + 1);
+  }
+ 
+  action start_keyword {
+    if (lexer->mark == 0) {
+      MARK(mark, p);
+    }
+  }
+  
+  action end_keyword {
+    MARK(keyword_end, p);
+    MARK(keyword_start, PTR_TO(mark));
+    MARK(content_start, p + 1);
+    lexer->mark = 0;
+  }
+  
+  action next_keyword_start {
+    MARK(content_end, p);
+  }
+
+  action start_row {
+    p = p - 1;
+    lexer->current_line = lexer->line_number;
+    current_row = rb_ary_new();
+  }
+  
+  action begin_cell_content {
+		MARK(content_start, p);
+  }
+
+  action store_cell_content {
+    VALUE con = ENCODED_STR_NEW(PTR_TO(content_start), LEN(content_start, p));
+    rb_funcall(con, rb_intern("strip!"), 0);
+    VALUE re_pipe      = rb_reg_regcomp(rb_str_new2("\\\\\\|"));
+    VALUE re_backslash = rb_reg_regcomp(rb_str_new2("\\\\\\\\"));
+    rb_funcall(con, rb_intern("gsub!"), 2, re_pipe,      rb_str_new2("|"));
+    rb_funcall(con, rb_intern("gsub!"), 2, re_backslash, rb_str_new2("\\"));
+
+    rb_ary_push(current_row, con);
+  }
+
+  action store_row {
+    rb_funcall(listener, rb_intern("row"), 2, current_row, INT2FIX(lexer->current_line));
+  }
+
+  action end_feature {
+    if (cs < lexer_first_final) {
+      if (raise_lexer_error != NULL) {
+        size_t count = 0;
+        int newstr_count = 0;
+        size_t len;
+        const char *buff;
+        if (lexer->last_newline != 0) {
+          len = LEN(last_newline, eof);
+          buff = PTR_TO(last_newline);
+        } else {
+          len = strlen(data);
+          buff = data;
+        }
+
+        char newstr[len]; 
+
+        for (count = 0; count < len; count++) {
+          if(buff[count] == 10) {
+            newstr[newstr_count] = '\0'; // terminate new string at first newline found
+            break;
+          } else {
+            if (buff[count] == '%') {
+              newstr[newstr_count++] = buff[count];
+              newstr[newstr_count] = buff[count];
+            } else {
+              newstr[newstr_count] = buff[count];
+            }
+          }
+          newstr_count++;
+        }
+
+        int line = lexer->line_number;
+        lexer_init(lexer); // Re-initialize so we can scan again with the same lexer
+        raise_lexer_error(newstr, line);
+      }
+    } else {
+      rb_funcall(listener, rb_intern("eof"), 0);
+    }
+  }
+
+  include lexer_common "lexer_common.<%= @i18n.underscored_iso_code %>.rl";   
+
+}%%
+
+/** Data **/
+%% write data;
+
+static VALUE 
+strip_i(VALUE str, VALUE ary)
+{
+  rb_funcall(str, rb_intern("strip!"), 0);
+  rb_ary_push(ary, str);
+  
+  return Qnil;
+}
+
+static VALUE 
+multiline_strip(VALUE text, int eol)
+{
+  VALUE map = rb_ary_new();
+  VALUE split = rb_str_split(text, "\n");
+  
+  rb_iterate(rb_each, split, strip_i, map);
+  
+  return rb_ary_join(split, rb_str_new2( \
+          eol == CRLF_FLAG ? CRLF : LF ));
+}
+
+static void 
+store_kw_con(VALUE listener, const char * event_name, 
+             const char * keyword_at, size_t keyword_length, 
+             const char * at,         size_t length, 
+             int current_line, int eol)
+{
+  VALUE con = Qnil, kw = Qnil;
+  kw = ENCODED_STR_NEW(keyword_at, keyword_length);
+  con = ENCODED_STR_NEW(at, length);
+  con = multiline_strip(con, eol);
+  rb_funcall(con, rb_intern("strip!"), 0);
+  rb_funcall(listener, rb_intern(event_name), 3, kw, con, INT2FIX(current_line)); 
+}
+
+static void 
+store_attr(VALUE listener, const char * attr_type,
+           const char * at, size_t length, 
+           int line)
+{
+  VALUE val = ENCODED_STR_NEW(at, length);
+  rb_funcall(listener, rb_intern(attr_type), 2, val, INT2FIX(line));
+}
+
+static void 
+store_pystring_content(VALUE listener, 
+          int start_col, 
+          const char *at, size_t length, 
+          int current_line)
+{
+  VALUE con = ENCODED_STR_NEW(at, length);
+  // Gherkin will crash gracefully if the string representation of start_col pushes the pattern past 32 characters
+  char pat[32]; 
+  snprintf(pat, 32, "^[\t ]{0,%d}", start_col); 
+  VALUE re = rb_reg_regcomp(rb_str_new2(pat));
+  VALUE re2 = rb_reg_regcomp(rb_str_new2("\r\\Z"));
+  VALUE unescape_escaped_quotes = rb_reg_regcomp(rb_str_new2("\\\\\"\\\\\"\\\\\""));
+  rb_funcall(con, rb_intern("gsub!"), 2, re, rb_str_new2(""));
+  rb_funcall(con, rb_intern("sub!"), 2, re2, rb_str_new2(""));
+  rb_funcall(con, rb_intern("gsub!"), 2, unescape_escaped_quotes, rb_str_new2("\"\"\""));
+  rb_funcall(listener, rb_intern("py_string"), 2, con, INT2FIX(current_line));
+}
+
+static void 
+raise_lexer_error(const char * at, int line)
+{ 
+  rb_raise(rb_eGherkinLexingError, "Lexing error on line %d: '%s'.", line, at);
+}
+
+static int 
+count_char(char char_to_count, char *str) {
+
+  int count = 0;
+  int i = 0;
+  while(str[i] != '\0') {
+    if(str[i] == char_to_count) {
+      count++;
+    }
+    i++;
+  }
+  return count;
+}
+
+static void lexer_init(lexer_state *lexer) {
+  lexer->content_start = 0;
+  lexer->content_end = 0;
+  lexer->content_len = 0;
+  lexer->mark = 0;
+  lexer->keyword_start = 0;
+  lexer->keyword_end = 0;
+  lexer->next_keyword_start = 0;
+  lexer->line_number = 1;
+  lexer->last_newline = 0;
+  lexer->final_newline = 0;
+  lexer->start_col = 0;
+  lexer->eol = LF_FLAG;
+}
+
+static VALUE CLexer_alloc(VALUE klass)
+{
+  VALUE obj;
+  lexer_state *lxr = ALLOC(lexer_state);
+  lexer_init(lxr);
+
+  obj = Data_Wrap_Struct(klass, NULL, -1, lxr);
+
+  return obj;
+}
+
+static VALUE CLexer_init(VALUE self, VALUE listener)
+{
+  rb_iv_set(self, "@listener", listener);
+  
+  lexer_state *lxr = NULL;
+  DATA_GET(self, lexer_state, lxr);
+  lexer_init(lxr);
+  
+  return self;
+}
+
+static VALUE CLexer_scan(VALUE self, VALUE input)
+{
+  lexer_state *lexer = NULL;
+  DATA_GET(self, lexer_state, lexer);
+
+
+  VALUE input_copy = rb_str_dup(input);
+
+  rb_str_append(input_copy, rb_str_new2("\n%_FEATURE_END_%"));
+  char *data = RSTRING_PTR(input_copy);
+  size_t len = RSTRING_LEN(input_copy);
+  
+  if (count_char('\r', data) > (count_char('\n', data) / 2)) {
+    lexer->eol = CRLF_FLAG;
+  }
+  
+  if (len == 0) { 
+    rb_raise(rb_eGherkinLexingError, "No content to lex.");
+  } else {
+
+    const char *p, *pe, *eof;
+    int cs = 0;
+    
+    VALUE listener = rb_iv_get(self, "@listener");
+    VALUE current_row = Qnil;
+
+    p = data;
+    pe = data + len;
+    eof = pe;
+    
+    assert(*pe == '\0' && "pointer does not end on NULL");
+    
+    %% write init;
+    %% write exec;
+
+    assert(p <= pe && "data overflow after parsing execute");
+    assert(lexer->content_start <= len && "content starts after data end");
+    assert(lexer->mark < len && "mark is after data end");
+    
+    // Reset lexer by re-initializing the whole thing
+    lexer_init(lexer);
+
+    if (cs == lexer_error) {
+      rb_raise(rb_eGherkinLexingError, "Invalid format, lexing fails.");
+    } else {
+      return Qtrue;
+    }
+  }
+}
+
+void Init_gherkin_lexer_<%= @i18n.underscored_iso_code %>()
+{
+  mGherkin = rb_define_module("Gherkin");
+  rb_eGherkinLexingError = rb_const_get(mGherkin, rb_intern("LexingError"));
+
+  mCLexer = rb_define_module_under(mGherkin, "CLexer");
+  cI18nLexer = rb_define_class_under(mCLexer, "<%= @i18n.underscored_iso_code.capitalize %>", rb_cObject);
+  rb_define_alloc_func(cI18nLexer, CLexer_alloc);
+  rb_define_method(cI18nLexer, "initialize", CLexer_init, 1);
+  rb_define_method(cI18nLexer, "scan", CLexer_scan, 1);
+}
+