regexp_parser 2.7.0 → 2.12.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 04af46818e9d560362fea9b3fd24802b557ac145ed95f6e02580dd7cf5e8ddfc
-  data.tar.gz: 75b7d30241f48ddf90c8cd68228fa928904ab6055ea755f4bdcf28361e645a4b
+  metadata.gz: aa5734a20a0705226c021d9a0efef48a6ca24b5b18f15f93b34c12ffe5021d1e
+  data.tar.gz: fa77ac98b3bd17d6bdba43d387ebbbd84a6a0a5fa86123c7eee68646e90ce37c
 SHA512:
-  metadata.gz: 407025a9b14af76463260fca2a48f9fef4ab863e3dddf3f7f54101c1348611afa49d9973e850d9e1c84d6e5faf8f1a9d3d2da5dceaefe8dc4fefe7069ecd9280
-  data.tar.gz: 9f3d2eb4264318511a82e9034c4c4a8a8e73e67e427945f0c9f745fd37b2f2f0ae8e30ba942f0920da3109b59436a5518dfc5e2f7669317de0214a0deb6f0e07
+  metadata.gz: 4bb62063aaa64e3828c5abf551fbf5ce83f726b5ebeb56af3e9210fd7e8c110c7f8f9ba1fcafd0dd0f66b8e8f36f04c1d72e5a739c4ace8fa19978dfc1a794b1
+  data.tar.gz: a82384912534ec3ca98d7665434c655468b15e4d75d6c1f13405f0b777d818746c6686d36049a158f0df9b829c129026487e5c3e6cf8a8f7c7ca65155a937db3

data/Gemfile

@@ -1,15 +1,17 @@
+# frozen_string_literal: true
+
 source 'https://rubygems.org'
 
 gemspec
 
 group :development, :test do
-  gem 'ice_nine', '~> 0.11.2'
-  gem 'rake', '~> 13.0'
-  gem 'regexp_property_values', '~> 1.3'
+  gem 'leto', '~> 2.1'
+  gem 'rake', '~> 13.1'
+  gem 'regexp_property_values', '~> 1.5'
   gem 'rspec', '~> 3.10'
   if RUBY_VERSION.to_f >= 2.7
     gem 'benchmark-ips', '~> 2.1'
-    gem 'gouteur'
-    gem 'rubocop', '~> 1.7'
+    gem 'gouteur', '~> 1.1'
+    gem 'rubocop', '>= 1.80.2'
   end
 end

data/LICENSE

@@ -1,4 +1,4 @@
-Copyright (c) 2010, 2012-2022,  Ammar Ali
+Copyright (c) 2010, 2012-2025,  Ammar Ali
 
 Permission is hereby granted, free of charge, to any person
 obtaining a copy of this software and associated documentation

data/Rakefile

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 require 'bundler'
 require 'rubygems'
 require 'rubygems/package_task'
@@ -14,10 +16,10 @@ RSpec::Core::RakeTask.new(:spec)
 task :default => [:'test:full']
 
 namespace :test do
-  task full: [:'ragel:rb', :spec]
+  task full: [:ragel, :spec]
 end
 
 # Add ragel task as a prerequisite for building the gem to ensure that the
 # latest scanner code is generated and included in the build.
-desc "Runs ragel:rb before building the gem"
-task :build => ['ragel:rb']
+desc "Runs ragel before building the gem"
+task build: :ragel

data/lib/regexp_parser/error.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 class Regexp::Parser
   # base class for all gem-specific errors
   class Error < StandardError; end

data/lib/regexp_parser/expression/base.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   class Base
     include Regexp::Expression::Shared
@@ -6,13 +8,6 @@ module Regexp::Expression
       init_from_token_and_options(token, options)
     end
 
-    def initialize_copy(orig)
-      self.text       = orig.text.dup         if orig.text
-      self.options    = orig.options.dup      if orig.options
-      self.quantifier = orig.quantifier.clone if orig.quantifier
-      super
-    end
-
     def to_re(format = :full)
       if set_level > 0
         warn "Calling #to_re on character set members is deprecated - "\

data/lib/regexp_parser/expression/classes/alternation.rb

@@ -1,5 +1,7 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
-  # A sequence of expressions, used by Alternation as one of its alternative.
+  # A sequence of expressions, used by Alternation as one of its alternatives.
   class Alternative < Regexp::Expression::Sequence; end
 
   class Alternation < Regexp::Expression::SequenceOperation

data/lib/regexp_parser/expression/classes/anchor.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   module Anchor
     class Base < Regexp::Expression::Base; end

data/lib/regexp_parser/expression/classes/backreference.rb

@@ -1,37 +1,15 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
-  # TODO: unify name with token :backref, one way or the other, in v3.0.0
   module Backreference
-    class Base < Regexp::Expression::Base
-      attr_accessor :referenced_expression
-
-      def initialize_copy(orig)
-        exp_id = [self.class, self.starts_at]
-
-        # prevent infinite recursion for recursive subexp calls
-        copied = @@copied ||= {}
-        self.referenced_expression =
-          if copied[exp_id]
-            orig.referenced_expression
-          else
-            copied[exp_id] = true
-            orig.referenced_expression.dup
-          end
-        copied.clear
-
-        super
-      end
-
-      def referential?
-        true
-      end
-    end
+    class Base < Regexp::Expression::Base; end
 
     class Number < Backreference::Base
       attr_reader :number
       alias reference number
 
       def initialize(token, options = {})
-        @number = token.text[token.token.equal?(:number) ? 1..-1 : 3..-2].to_i
+        @number = token.text[/-?\d+/].to_i
         super
       end
     end
@@ -74,4 +52,7 @@ module Regexp::Expression
       end
     end
   end
+
+  # alias for symmetry between token symbol and Expression class name
+  Backref = Backreference
 end

data/lib/regexp_parser/expression/classes/character_set/intersection.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   class CharacterSet < Regexp::Expression::Subexpression
     class IntersectedSequence < Regexp::Expression::Sequence; end

data/lib/regexp_parser/expression/classes/character_set/range.rb

@@ -1,10 +1,11 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   class CharacterSet < Regexp::Expression::Subexpression
     class Range < Regexp::Expression::Subexpression
-      def starts_at
-        expressions.first.starts_at
+      def ts
+        (head = expressions.first) ? head.ts : @ts
       end
-      alias :ts :starts_at
 
       def <<(exp)
         complete? and raise Regexp::Parser::Error,
@@ -15,10 +16,6 @@ module Regexp::Expression
       def complete?
         count == 2
       end
-
-      def parts
-        intersperse(expressions, text.dup)
-      end
     end
   end
 end

data/lib/regexp_parser/expression/classes/character_set.rb

@@ -1,10 +1,9 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   class CharacterSet < Regexp::Expression::Subexpression
     attr_accessor :closed, :negative
-
-    alias :negative? :negative
-    alias :negated?  :negative
-    alias :closed?   :closed
+    alias :closed? :closed
 
     def initialize(token, options = {})
       self.negative = false
@@ -19,9 +18,8 @@ module Regexp::Expression
     def close
       self.closed = true
     end
-
-    def parts
-      ["#{text}#{'^' if negated?}", *expressions, ']']
-    end
   end
+
+  # alias for symmetry between token symbol and Expression class name
+  Set = CharacterSet
 end # module Regexp::Expression

data/lib/regexp_parser/expression/classes/character_type.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   module CharacterType
     class Base < Regexp::Expression::Base; end

data/lib/regexp_parser/expression/classes/conditional.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   module Conditional
     class TooManyBranches < Regexp::Parser::Error
@@ -7,37 +9,24 @@ module Regexp::Expression
     end
 
     class Condition < Regexp::Expression::Base
-      attr_accessor :referenced_expression
-
       # Name or number of the referenced capturing group that determines state.
       # Returns a String if reference is by name, Integer if by number.
       def reference
         ref = text.tr("'<>()", "")
         ref =~ /\D/ ? ref : Integer(ref)
       end
-
-      def initialize_copy(orig)
-        self.referenced_expression = orig.referenced_expression.dup
-        super
-      end
-
-      def referential?
-        true
-      end
     end
 
     class Branch < Regexp::Expression::Sequence; end
 
     class Expression < Regexp::Expression::Subexpression
-      attr_accessor :referenced_expression
-
       def <<(exp)
         expressions.last << exp
       end
 
-      def add_sequence(active_opts = {})
+      def add_sequence(active_opts = {}, params = { ts: 0 })
         raise TooManyBranches.new if branches.length == 2
-        params = { conditional_level: conditional_level + 1 }
+        params = params.merge({ conditional_level: conditional_level + 1 })
         Branch.add_to(self, params, active_opts)
       end
       alias :branch :add_sequence
@@ -58,19 +47,6 @@ module Regexp::Expression
       def reference
         condition.reference
       end
-
-      def referential?
-        true
-      end
-
-      def parts
-        [text.dup, condition, *intersperse(branches, '|'), ')']
-      end
-
-      def initialize_copy(orig)
-        self.referenced_expression = orig.referenced_expression.dup
-        super
-      end
     end
   end
 end

data/lib/regexp_parser/expression/classes/escape_sequence.rb

@@ -1,100 +1,33 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
-  # TODO: unify naming with Token::Escape, one way or the other, in v3.0.0
   module EscapeSequence
-    class Base < Regexp::Expression::Base
-      def codepoint
-        char.ord
-      end
-
-      if ''.respond_to?(:undump)
-        def char
-          %("#{text}").undump
-        end
-      else
-        # poor man's unescape without using eval
-        require 'yaml'
-        def char
-          YAML.load(%Q(---\n"#{text}"\n))
-        end
-      end
-    end
-
-    class Literal < EscapeSequence::Base
-      def char
-        text[1..-1]
-      end
-    end
-
-    class AsciiEscape   < EscapeSequence::Base; end
-    class Backspace     < EscapeSequence::Base; end
-    class Bell          < EscapeSequence::Base; end
-    class FormFeed      < EscapeSequence::Base; end
-    class Newline       < EscapeSequence::Base; end
-    class Return        < EscapeSequence::Base; end
-    class Tab           < EscapeSequence::Base; end
-    class VerticalTab   < EscapeSequence::Base; end
-
-    class Hex           < EscapeSequence::Base; end
-    class Codepoint     < EscapeSequence::Base; end
-
-    class CodepointList < EscapeSequence::Base
-      def char
-        raise NoMethodError, 'CodepointList responds only to #chars'
-      end
-
-      def codepoint
-        raise NoMethodError, 'CodepointList responds only to #codepoints'
-      end
-
-      def chars
-        codepoints.map { |cp| cp.chr('utf-8') }
-      end
-
-      def codepoints
-        text.scan(/\h+/).map(&:hex)
-      end
-    end
-
-    class Octal < EscapeSequence::Base
-      def char
-        text[1..-1].to_i(8).chr('utf-8')
-      end
-    end
-
-    class AbstractMetaControlSequence < EscapeSequence::Base
-      def char
-        codepoint.chr('utf-8')
-      end
-
-      private
-
-      def control_sequence_to_s(control_sequence)
-        five_lsb = control_sequence.unpack('B*').first[-5..-1]
-        ["000#{five_lsb}"].pack('B*')
-      end
-
-      def meta_char_to_codepoint(meta_char)
-        byte_value = meta_char.ord
-        byte_value < 128 ? byte_value + 128 : byte_value
-      end
-    end
-
-    class Control < AbstractMetaControlSequence
-      def codepoint
-        control_sequence_to_s(text).ord
-      end
-    end
-
-    class Meta < AbstractMetaControlSequence
-      def codepoint
-        meta_char_to_codepoint(text[-1])
-      end
-    end
-
-    class MetaControl < AbstractMetaControlSequence
-      def codepoint
-        meta_char_to_codepoint(control_sequence_to_s(text))
-      end
-    end
+    Base        = Class.new(Regexp::Expression::Base)
+
+    AsciiEscape = Class.new(Base) # \e
+    Backspace   = Class.new(Base) # \b
+    Bell        = Class.new(Base) # \a
+    FormFeed    = Class.new(Base) # \f
+    Newline     = Class.new(Base) # \n
+    Return      = Class.new(Base) # \r
+    Tab         = Class.new(Base) # \t
+    VerticalTab = Class.new(Base) # \v
+
+    Literal     = Class.new(Base) # e.g. \j, \@, \😀 (ineffectual escapes)
+
+    Octal       = Class.new(Base) # e.g. \012
+    Hex         = Class.new(Base) # e.g. \x0A
+    Codepoint   = Class.new(Base) # e.g. \u000A
+
+    CodepointList = Class.new(Base) # e.g. \u{A B}
+    UTF8Hex       = Class.new(Base) # e.g. \xE2\x82\xAC
+
+    AbstractMetaControlSequence = Class.new(Base)
+    Control                     = Class.new(AbstractMetaControlSequence) # e.g. \cB
+    Meta                        = Class.new(AbstractMetaControlSequence) # e.g. \M-Z
+    MetaControl                 = Class.new(AbstractMetaControlSequence) # e.g. \M-\cX
   end
+
+  # alias for symmetry between Token::* and Expression::*
+  Escape = EscapeSequence
 end

data/lib/regexp_parser/expression/classes/free_space.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   class FreeSpace < Regexp::Expression::Base
     def quantify(*_args)
@@ -5,10 +7,12 @@ module Regexp::Expression
     end
   end
 
-  class Comment < Regexp::Expression::FreeSpace; end
+  class Comment < Regexp::Expression::FreeSpace
+  end
 
   class WhiteSpace < Regexp::Expression::FreeSpace
     def merge(exp)
+      warn("#{self.class}##{__method__} is deprecated and will be removed in v3.0.0.")
       text << exp.text
     end
   end

data/lib/regexp_parser/expression/classes/group.rb

@@ -1,13 +1,8 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   module Group
     class Base < Regexp::Expression::Subexpression
-      def parts
-        [text.dup, *expressions, ')']
-      end
-
-      def capturing?; false end
-
-      def comment?; false end
     end
 
     class Passive < Group::Base
@@ -18,14 +13,6 @@ module Regexp::Expression
         super
       end
 
-      def parts
-        if implicit?
-          expressions
-        else
-          super
-        end
-      end
-
       def implicit?
         @implicit
       end
@@ -55,8 +42,6 @@ module Regexp::Expression
     class Capture < Group::Base
       attr_accessor :number, :number_at_level
       alias identifier number
-
-      def capturing?; true end
     end
 
     class Named < Group::Capture
@@ -75,11 +60,6 @@ module Regexp::Expression
     end
 
     class Comment < Group::Base
-      def parts
-        [text.dup]
-      end
-
-      def comment?; true end
     end
   end
 

data/lib/regexp_parser/expression/classes/keep.rb

@@ -1,6 +1,8 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   module Keep
-    # TOOD: in regexp_parser v3.0.0 this should possibly be a Subexpression
+    # TODO: in regexp_parser v3.0.0 this should possibly be a Subexpression
     #       that contains all expressions to its left.
     class Mark < Regexp::Expression::Base; end
   end

data/lib/regexp_parser/expression/classes/literal.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   class Literal < Regexp::Expression::Base; end
 end

data/lib/regexp_parser/expression/classes/posix_class.rb

@@ -1,11 +1,13 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   class PosixClass < Regexp::Expression::Base
-    def negative?
-      type == :nonposixclass
-    end
-
     def name
-      token.to_s
+      text[/\w+/]
     end
   end
+
+  # alias for symmetry between token symbol and Expression class name
+  Posixclass    = PosixClass
+  Nonposixclass = PosixClass
 end

data/lib/regexp_parser/expression/classes/root.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   class Root < Regexp::Expression::Subexpression
     def self.build(options = {})

data/lib/regexp_parser/expression/classes/unicode_property.rb

@@ -1,17 +1,14 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
-  # TODO: unify name with token :property, one way or the other, in v3.0.0
   module UnicodeProperty
     class Base < Regexp::Expression::Base
-      def negative?
-        type == :nonproperty
-      end
-
       def name
         text[/\A\\[pP]\{([^}]+)\}\z/, 1]
       end
 
       def shortcut
-        (Regexp::Scanner.short_prop_map.rassoc(token.to_s) || []).first
+        Regexp::Scanner.short_prop_map.key(token.to_s)
       end
     end
 
@@ -110,10 +107,15 @@ module Regexp::Expression
       class Unassigned  < Codepoint::Base; end
     end
 
-    class Age     < UnicodeProperty::Base; end
-    class Derived < UnicodeProperty::Base; end
-    class Emoji   < UnicodeProperty::Base; end
-    class Script  < UnicodeProperty::Base; end
-    class Block   < UnicodeProperty::Base; end
+    class Age        < UnicodeProperty::Base; end
+    class Block      < UnicodeProperty::Base; end
+    class Derived    < UnicodeProperty::Base; end
+    class Emoji      < UnicodeProperty::Base; end
+    class Enumerated < UnicodeProperty::Base; end
+    class Script     < UnicodeProperty::Base; end
   end
+
+  # alias for symmetry between token symbol and Expression class name
+  Property    = UnicodeProperty
+  Nonproperty = UnicodeProperty
 end # module Regexp::Expression

data/lib/regexp_parser/expression/methods/construct.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   module Shared
     module ClassMethods
@@ -25,11 +27,9 @@ module Regexp::Expression
       def token_class
         if self == Root || self < Sequence
           nil # no token class because these objects are Parser-generated
-        # TODO: synch exp & token class names for alt., dot, escapes in v3.0.0
-        elsif self == Alternation || self == CharacterType::Any
+        # TODO: synch exp class, token class & type names for this in v3.0.0
+        elsif self == CharacterType::Any
           Regexp::Syntax::Token::Meta
-        elsif self <= EscapeSequence::Base
-          Regexp::Syntax::Token::Escape
         else
           Regexp::Syntax::Token.const_get(name.split('::')[2])
         end

data/lib/regexp_parser/expression/methods/escape_sequence_char.rb

@@ -0,0 +1,7 @@
+# frozen_string_literal: true
+
+Regexp::Expression::EscapeSequence::Base.class_eval do
+  def char
+    codepoint.chr('utf-8')
+  end
+end