srl_ruby 0.2.5 → 0.2.6

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 13bb293008059e97968eab3b976e05f7fe9c353e
-  data.tar.gz: 933e9531ae294fe4adb46e3c7ef87c58a3243388
+  metadata.gz: 475a892be14d441ff82714324f489707839eae57
+  data.tar.gz: 52ad8fffe6fb8abe17e99e690dc182bb0b3a48b0
 SHA512:
-  metadata.gz: 1c585bfad3f6330b87f47ab7c1dae115bc953a50220bc3d87642524c523cfbe1ae19193cfc0173730d456ec6691611abd59964ea4722a42bdbb1571f6c06ce9c
-  data.tar.gz: f835d81332961e020addee4ec8b5e4c46378f29ad12a7497fb03f5545eee1c884b1da5b8a621d4f76713b348b3b324fc96c4eb9f60b98037c1c7e9b98ab5d9c6
+  metadata.gz: 24b028b5becb5f2ac57c6fbf1aaa5eb45d7b3e3bde1e8ea2892b56f7d67fc643bf8d28cff0e431e7a3f97c996b0c44321bfd52eb17d818daeb51a334cb2d056b
+  data.tar.gz: b21b2fefc35eac997bd06a0a4a7bc586d80eba2074cfb365c7a22a4e6390e48a45960df3132b09abc2e53569c5c6ad4f925510191808212ba6a8b991dfc259a8

data/CHANGELOG.md

@@ -1,5 +1,14 @@
+## [0.2.6] - 2018-04-03
+SrlRuby passes 13 tests out of 15 from standard SRL test suite. 
+### Changed
+- Class `SrlRuby#Tokenizer` added CHAR_CLASS literal and keywords EITHER, NONE, EITHER,
+- Grammar expanded to support 'CARRIAGE RETURN', 'VERTICAL TAB', 'WORD', 'NO WORD' expressions
+- Class `SrlRuby::ASTBuilder` updates to reflect changes in the grammar.
+- File `acceptance/srl_test_suite_spec.rb`. 13 test files from official test suite are passing.
+
+
 ## [0.2.5] - 2018-04-02
-SrlRuby passes 12 tests out of 15 standard SRL tests in total. 
+SrlRuby passes 12 tests out of 15 from standard SRL test suite. 
 ### Changed
 - Class `SrlRuby#Tokenizer` added keywords CARRIAGE, RETURN, VERTICAL, WORD
 - Grammar expanded to support 'CARRIAGE RETURN', 'VERTICAL TAB', 'WORD', 'NO WORD' expressions
@@ -8,7 +17,7 @@ SrlRuby passes 12 tests out of 15 standard SRL tests in total.
 
 
 ## [0.2.4] - 2018-04-02
-SrlRuby passes 10 tests out of 15 standard SRL tests in total. 
+SrlRuby passes 10 tests out of 15 from standard SRL test suite. 
 ### Changed
 - File `lib/srl_ruby/grammar.rb` grammar refactoring. Added support for new 'no digit' SRL expression.
 - Class `SrlRuby::ASTBuilder` updates to reflect changes in the grammar.
@@ -27,7 +36,7 @@ SrlRuby passes 10 tests out of 15 standard SRL tests in total.
 - Fixed rley version dependency
 
 ## [0.2.1] - 2018-03-15
-SrlRuby passes 7 tests out of 15 standard SRL tests in total.  
+SrlRuby passes 7 tests out of 15 from standard SRL test suite.  
 ### Changed
 - File `acceptance/srl_test_suite_spec.rb`. More examples in spec file.
 - File `ast_builder.rb` updates to reflect grammar changes.
@@ -39,7 +48,7 @@ SrlRuby passes 7 tests out of 15 standard SRL tests in total.
 - File `ast_builder.rb` fixed anchor implementation.
 
 ## [0.2.0] - 2018-03-14
-SrlRuby passes 3 standard out of 15 standard SRL tests in total.
+SrlRuby passes 3 tests out of 15 from standard SRL test suite.
 ### Added
 - Added `spec/acceptance/support` directory. It contains test harness to use the .rule files from standard SRL test suite. 
 - Added `acceptance/srl_test_suite_spec.rb`file. Spec file designed to standard SRL test suite. At this date, SrlRuby passes 3 tests out of 15 tests in total.

data/lib/regex/raw_expression.rb

@@ -0,0 +1,23 @@
+require_relative 'atomic_expression' # Access the superclass
+
+module Regex # This module is used as a namespace
+  # A raw expression is a string that will be copied verbatim (as is)
+  # in the generated regular expression.
+  class RawExpression < AtomicExpression
+	attr_reader :raw
+  
+    # Constructor
+    def initialize(rawLiteral)
+      super()
+      @raw = rawLiteral
+    end
+
+    protected
+
+    # Conversion method re-definition.
+    # Purpose: Return the String representation of the expression.
+	alias_method :text_repr, :raw
+  end # class
+end # module
+
+# End of file

data/lib/srl_ruby/ast_builder.rb

@@ -272,15 +272,24 @@ module SrlRuby
       return Regex::CharClass.new(false, *alternatives)
     end
 
+    # rule('character_class' => %w[NONE OF STRING_LIT]).as 'none_of'
+    def reduce_none_of(_production, _range, _tokens, theChildren)
+      raw_literal = theChildren[-1].token.lexeme.dup
+      chars = raw_literal.chars.map do |ch|
+        Regex::Character.new(ch)
+      end
+      Regex::CharClass.new(true, *chars)
+    end
+
     # rule('special_char' => 'TAB').as 'tab'
     def reduce_tab(_production, _range, _tokens, _children)
       Regex::Character.new('\t')
     end
-    
+
     # rule('special_char' => ' VERTICAL TAB').as 'vtab'
     def reduce_vtab(_production, _range, _tokens, _children)
       Regex::Character.new('\v')
-    end    
+    end
 
     # rule('special_char' => 'BACKSLASH').as 'backslash'
     def reduce_backslash(_production, _range, _tokens, _children)
@@ -293,7 +302,7 @@ module SrlRuby
       # TODO: control portability
       Regex::Character.new('\n')
     end
-    
+
     # rule('special_char' => %w[CARRIAGE RETURN]).as 'carriage_return'
     def reduce_carriage_return(_production, _range, _tokens, _children)
       Regex::Character.new('\r')
@@ -303,11 +312,11 @@ module SrlRuby
     def reduce_word(_production, _range, _tokens, _children)
       Regex::Anchor.new('\b')
     end
-    
+
     # rule('special_char' => %w[NO WORD]).as 'no word'
     def reduce_no_word(_production, _range, _tokens, _children)
       Regex::Anchor.new('\B')
-    end    
+    end
 
     # rule('literal' => %w[LITERALLY STRING_LIT]).as 'literally'
     def reduce_literally(_production, _range, _tokens, theChildren)
@@ -317,6 +326,12 @@ module SrlRuby
       return string_literal(raw_literal)
     end
 
+    # rule('raw' => %w[RAW STRING_LIT]).as 'raw_literal'
+    def reduce_raw_literal(_production, _range, _tokens, theChildren)
+      raw_literal = theChildren[-1].token.lexeme.dup
+      return Regex::RawExpression.new(raw_literal)
+    end
+
     # rule('alternation' => %w[ANY OF LPAREN alternatives RPAREN]).as 'any_of'
     def reduce_any_of(_production, _range, _tokens, theChildren)
       first_alternative = theChildren[3].first

data/lib/srl_ruby/grammar.rb

@@ -4,17 +4,22 @@ module SrlRuby
   ########################################
   # SRL grammar
   builder = Rley::Syntax::GrammarBuilder.new do
+    # Separators...
     add_terminals('LPAREN', 'RPAREN', 'COMMA')
-    add_terminals('DIGIT_LIT', 'INTEGER', 'LETTER_LIT')
+    
+    # Literal values...
+    add_terminals('DIGIT_LIT', 'INTEGER', 'LETTER_LIT', 'CHAR_CLASS')
     add_terminals('LITERALLY', 'STRING_LIT', 'IDENTIFIER')
+    
+    # Keywords...
     add_terminals('BEGIN', 'STARTS', 'WITH')
-    add_terminals('MUST', 'END')
+    add_terminals('MUST', 'END', 'RAW')
     add_terminals('UPPERCASE', 'LETTER', 'FROM', 'TO')
-    add_terminals('DIGIT', 'NUMBER', 'ANY', 'NO')
+    add_terminals('DIGIT', 'NUMBER', 'ANY', 'EITHER', 'NO')
     add_terminals('CHARACTER', 'WHITESPACE', 'ANYTHING')
     add_terminals('TAB', 'BACKSLASH', 'NEW', 'LINE', 'WORD')
-    add_terminals('CARRIAGE', 'RETURN', 'VERTICAL', 'OF', 'ONE')
-    add_terminals('EXACTLY', 'TIMES', 'ONCE', 'TWICE')
+    add_terminals('CARRIAGE', 'RETURN', 'VERTICAL', 'OF', 'NONE')
+    add_terminals('ONE', 'EXACTLY', 'TIMES', 'ONCE', 'TWICE')
     add_terminals('BETWEEN', 'AND', 'OPTIONAL', 'OR')
     add_terminals('MORE', 'NEVER', 'AT', 'LEAST')
     add_terminals('IF', 'FOLLOWED', 'BY', 'NOT')
@@ -23,6 +28,7 @@ module SrlRuby
     add_terminals('CASE', 'INSENSITIVE', 'MULTI', 'ALL')
     add_terminals('LAZY')
 
+    # Grammar rules...
     rule('srl' => 'expression').as 'start_rule'
     rule('expression' => %w[pattern flags]).as 'flagged_expr'
     rule('expression' => 'pattern').as 'simple_expr'
@@ -60,6 +66,7 @@ module SrlRuby
     rule('atom' => 'character_class').as 'character_class_atom'
     rule('atom' => 'special_char').as 'special_char_atom'
     rule('atom' => 'literal').as 'literal_atom'
+    rule('atom' => 'raw').as 'raw_atom'
     rule('letter_range' => %w[LETTER FROM LETTER_LIT TO LETTER_LIT]).as 'lowercase_from_to'
     rule('letter_range' => %w[UPPERCASE LETTER FROM LETTER_LIT TO LETTER_LIT]).as 'uppercase_from_to'
     rule('letter_range' => 'LETTER').as 'any_lowercase'
@@ -72,7 +79,12 @@ module SrlRuby
     rule('character_class' => 'WHITESPACE').as 'whitespace'
     rule('character_class' => %w[NO WHITESPACE]).as 'no_whitespace'
     rule('character_class' => 'ANYTHING').as 'anything'
-    rule('character_class' => %w[ONE OF STRING_LIT]).as 'one_of'
+    rule('character_class' => %w[ONE OF cclass]).as 'one_of'
+    rule('character_class' => %w[NONE OF cclass]).as 'none_of'
+    rule('cclass' => 'STRING_LIT').as 'quoted_cclass' # Preferred syntax
+    rule('cclass' => 'INTEGER').as 'digits_cclass'
+    rule('cclass' => 'IDENTIFIER').as 'identifier_cclass'
+    rule('cclass' => 'CHAR_CLASS').as 'unquoted_cclass'
     rule('special_char' => 'TAB').as 'tab'
     rule('special_char' => 'VERTICAL TAB').as 'vtab'
     rule('special_char' => 'BACKSLASH').as 'backslash'
@@ -81,9 +93,12 @@ module SrlRuby
     rule('special_char' => %w[WORD]).as 'word'
     rule('special_char' => %w[NO WORD]).as 'no_word'
     rule('literal' => %w[LITERALLY STRING_LIT]).as 'literally'
-    rule('alternation' => %w[ANY OF LPAREN alternatives RPAREN]).as 'any_of'
+    rule('raw' => 'RAW STRING_LIT').as 'raw_literal'
+    rule('alternation' => %w[any_or_either OF LPAREN alternatives RPAREN]).as 'any_of'
     rule('alternatives' => %w[alternatives separator quantifiable]).as 'alternative_list'
     rule('alternatives' => 'quantifiable').as 'simple_alternative'
+    rule('any_or_either' => 'ANY').as 'any_keyword'
+    rule('any_or_either' => 'EITHER').as 'either_keyword'
     rule('grouping' => %w[LPAREN pattern RPAREN]).as 'grouping_parenthenses'
     rule('capturing_group' => %w[CAPTURE assertable]).as 'capture'
     rule('capturing_group' => %w[CAPTURE assertable UNTIL assertable]).as 'capture_until'

data/lib/srl_ruby/regex_repr.rb

@@ -11,4 +11,5 @@ require_relative '../regex/non_capturing_group'
 require_relative '../regex/anchor'
 require_relative '../regex/lookaround'
 require_relative '../regex/capturing_group'
-require_relative '../regex/match_option'
+require_relative '../regex/match_option'
+require_relative '../regex/raw_expression'

data/lib/srl_ruby/tokenizer.rb

@@ -42,6 +42,7 @@ module SrlRuby
       CASE
       CHARACTER
       DIGIT
+      EITHER
       END
       EXACTLY
       FOLLOWED
@@ -60,6 +61,7 @@ module SrlRuby
       NEVER
       NEW
       NO
+      NONE
       NOT
       NUMBER
       OF
@@ -67,6 +69,7 @@ module SrlRuby
       ONE
       OPTIONAL
       OR
+      RAW
       RETURN
       STARTS
       TAB
@@ -111,22 +114,24 @@ module SrlRuby
       if '(),'.include? curr_ch
         # Delimiters, separators => single character token
         token = build_token(@@lexeme2name[curr_ch], scanner.getch)
-      elsif (lexeme = scanner.scan(/[0-9]{2,}/))
+      elsif (lexeme = scanner.scan(/[0-9]{2,}((?=\s)|$)/))
         token = build_token('INTEGER', lexeme) # An integer has 2..* digits
-      elsif (lexeme = scanner.scan(/[0-9]/))
+      elsif (lexeme = scanner.scan(/[0-9]((?=\s)|$)/))
         token = build_token('DIGIT_LIT', lexeme)
-      elsif (lexeme = scanner.scan(/[a-zA-Z_][a-zA-Z0-9_]+/))
-        keyw = @@keywords[lexeme.upcase]
-        tok_type = keyw ? keyw : 'IDENTIFIER'
-        token = build_token(tok_type, lexeme)
-      elsif (lexeme = scanner.scan(/[a-zA-Z]((?=\s)|$)/))
-        token = build_token('LETTER_LIT', lexeme)
       elsif (lexeme = scanner.scan(/"(?:\\"|[^"])*"/)) # Double quotes literal?
         unquoted = lexeme.gsub(/(^")|("$)/, '')
         token = build_token('STRING_LIT', unquoted)
       elsif (lexeme = scanner.scan(/'(?:\\'|[^'])*'/)) # Single quotes literal?
         unquoted = lexeme.gsub(/(^')|('$)/, '')
         token = build_token('STRING_LIT', unquoted)
+      elsif (lexeme = scanner.scan(/[a-zA-Z]((?=\s)|$)/))        
+        token = build_token('LETTER_LIT', lexeme)
+      elsif (lexeme = scanner.scan(/[a-zA-Z_][a-zA-Z0-9_]+/))
+        keyw = @@keywords[lexeme.upcase]
+        tok_type = keyw ? keyw : 'IDENTIFIER'
+        token = build_token(tok_type, lexeme)       
+      elsif (lexeme = scanner.scan(/[^,"\s]{2,}/))
+        token = build_token('CHAR_CLASS', lexeme)
       else # Unknown token
         erroneous = curr_ch.nil? ? '' : curr_ch
         sequel = scanner.scan(/.{1,20}/)

data/lib/srl_ruby/version.rb

@@ -1,3 +1,3 @@
 module SrlRuby
-  VERSION = '0.2.5'.freeze
+  VERSION = '0.2.6'.freeze
 end

data/spec/acceptance/srl_test_suite_spec.rb

@@ -71,6 +71,11 @@ RSpec.describe Acceptance do
     rule_file_repr = load_file('nondigit.rule')
     test_rule_file(rule_file_repr)
   end
+  
+  it 'should support negative character class' do
+    rule_file_repr = load_file('none_of.rule')
+    test_rule_file(rule_file_repr)
+  end
 
   it 'should match a tab' do
     rule_file_repr = load_file('tab.rule')

data/spec/srl_ruby_spec.rb

@@ -44,6 +44,11 @@ describe SrlRuby do
       regexp = SrlRuby.parse("literally '.'")
       expect(regexp.source).to eq('\.')
     end
+    
+    it 'should parse single quotes literal string' do
+      regexp = SrlRuby.parse('literally "an", whitespace, raw "[a-zA-Z]"')
+      expect(regexp.source).to eq('an\s[a-zA-Z]')
+    end
   end # context
 
   context 'Parsing character classes:' do
@@ -93,6 +98,41 @@ describe SrlRuby do
       # (escapes more characters than required)
       expect(regexp.source).to eq('[._%+\-]')
     end
+    
+    it "should parse 'one of' with unquoted character class syntax" do
+      # Case of digit sequence
+      regexp = SrlRuby.parse('one of 13579, must end')
+      expect(regexp.source).to eq('[13579]$')
+      
+      # Case of identifier-like character class
+      regexp = SrlRuby.parse('one of abcd, must end')
+      expect(regexp.source).to eq('[abcd]$')
+
+      # Case of arbitrary character class
+      regexp = SrlRuby.parse('one of 12hms:, must end')
+      expect(regexp.source).to eq('[12hms:]$')       
+    end    
+    
+    it "should parse 'none of' syntax" do
+      regexp = SrlRuby.parse('none of "._%+-"')
+      # Remark: reference implementation less readable
+      # (escapes more characters than required)
+      expect(regexp.source).to eq('[^._%+\-]')
+    end 
+
+    it "should parse 'none of' with unquoted character class syntax" do
+      # Case of digit sequence
+      regexp = SrlRuby.parse('none of 13579, must end')
+      expect(regexp.source).to eq('[^13579]$')
+      
+      # Case of identifier-like character class
+      regexp = SrlRuby.parse('none of abcd, must end')
+      expect(regexp.source).to eq('[^abcd]$')
+
+      # Case of arbitrary character class
+      regexp = SrlRuby.parse('none of 12hms:^, must end')
+      expect(regexp.source).to eq('[^12hms:\^]$')       
+    end    
   end # context
 
   context 'Parsing special character declarations:' do
@@ -137,6 +177,12 @@ describe SrlRuby do
       source = 'any of (any character, one of "._%-+")'
       regexp = SrlRuby.parse(source)
       expect(regexp.source).to eq('(?:\w|[._%\-+])')
+    end 
+
+    it "should parse 'either of' syntax" do
+      source = 'either of (any character, one of "._%-+")'
+      regexp = SrlRuby.parse(source)
+      expect(regexp.source).to eq('(?:\w|[._%\-+])')
     end    
     
     it 'should anchor as alternative' do

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: srl_ruby
 version: !ruby/object:Gem::Version
-  version: 0.2.5
+  version: 0.2.6
 platform: ruby
 authors:
 - Dimitri Geshef
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2018-04-02 00:00:00.000000000 Z
+date: 2018-04-03 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rley
@@ -104,6 +104,7 @@ files:
 - lib/regex/non_capturing_group.rb
 - lib/regex/polyadic_expression.rb
 - lib/regex/quantifiable.rb
+- lib/regex/raw_expression.rb
 - lib/regex/repetition.rb
 - lib/regex/wildcard.rb
 - lib/srl_ruby.rb