regexp_parser 2.8.1 → 2.11.3

data/lib/regexp_parser/expression/shared.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   module Shared
     module ClassMethods; end # filled in ./methods/*.rb
@@ -70,11 +72,12 @@ module Regexp::Expression
     # lit.to_s(:original) # => 'a +' # with quantifier AND intermittent decorations
     #
     def to_s(format = :full)
-      base = parts.each_with_object(''.dup) do |part, buff|
+      base = ''.dup
+      parts.each do |part|
         if part.instance_of?(String)
-          buff << part
+          base << part
         elsif !part.custom_to_s_handling
-          buff << part.to_s(:original)
+          base << part.to_s(:original)
         end
       end
       "#{base}#{pre_quantifier_decoration(format)}#{quantifier_affix(format)}"

data/lib/regexp_parser/expression/subexpression.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 module Regexp::Expression
   class Subexpression < Regexp::Expression::Base
     include Enumerable

data/lib/regexp_parser/expression.rb

@@ -1,36 +1,42 @@
-require 'regexp_parser/error'
+# frozen_string_literal: true
 
-require 'regexp_parser/expression/shared'
-require 'regexp_parser/expression/base'
-require 'regexp_parser/expression/quantifier'
-require 'regexp_parser/expression/subexpression'
-require 'regexp_parser/expression/sequence'
-require 'regexp_parser/expression/sequence_operation'
+require_relative 'error'
 
-require 'regexp_parser/expression/classes/alternation'
-require 'regexp_parser/expression/classes/anchor'
-require 'regexp_parser/expression/classes/backreference'
-require 'regexp_parser/expression/classes/character_set'
-require 'regexp_parser/expression/classes/character_set/intersection'
-require 'regexp_parser/expression/classes/character_set/range'
-require 'regexp_parser/expression/classes/character_type'
-require 'regexp_parser/expression/classes/conditional'
-require 'regexp_parser/expression/classes/escape_sequence'
-require 'regexp_parser/expression/classes/free_space'
-require 'regexp_parser/expression/classes/group'
-require 'regexp_parser/expression/classes/keep'
-require 'regexp_parser/expression/classes/literal'
-require 'regexp_parser/expression/classes/posix_class'
-require 'regexp_parser/expression/classes/root'
-require 'regexp_parser/expression/classes/unicode_property'
+require_relative 'expression/shared'
+require_relative 'expression/base'
+require_relative 'expression/quantifier'
+require_relative 'expression/subexpression'
+require_relative 'expression/sequence'
+require_relative 'expression/sequence_operation'
 
-require 'regexp_parser/expression/methods/construct'
-require 'regexp_parser/expression/methods/human_name'
-require 'regexp_parser/expression/methods/match'
-require 'regexp_parser/expression/methods/match_length'
-require 'regexp_parser/expression/methods/options'
-require 'regexp_parser/expression/methods/parts'
-require 'regexp_parser/expression/methods/printing'
-require 'regexp_parser/expression/methods/strfregexp'
-require 'regexp_parser/expression/methods/tests'
-require 'regexp_parser/expression/methods/traverse'
+require_relative 'expression/classes/alternation'
+require_relative 'expression/classes/anchor'
+require_relative 'expression/classes/backreference'
+require_relative 'expression/classes/character_set'
+require_relative 'expression/classes/character_set/intersection'
+require_relative 'expression/classes/character_set/range'
+require_relative 'expression/classes/character_type'
+require_relative 'expression/classes/conditional'
+require_relative 'expression/classes/escape_sequence'
+require_relative 'expression/classes/free_space'
+require_relative 'expression/classes/group'
+require_relative 'expression/classes/keep'
+require_relative 'expression/classes/literal'
+require_relative 'expression/classes/posix_class'
+require_relative 'expression/classes/root'
+require_relative 'expression/classes/unicode_property'
+
+require_relative 'expression/methods/construct'
+require_relative 'expression/methods/escape_sequence_char'
+require_relative 'expression/methods/escape_sequence_codepoint'
+require_relative 'expression/methods/human_name'
+require_relative 'expression/methods/match'
+require_relative 'expression/methods/match_length'
+require_relative 'expression/methods/negative'
+require_relative 'expression/methods/options'
+require_relative 'expression/methods/parts'
+require_relative 'expression/methods/printing'
+require_relative 'expression/methods/referenced_expressions'
+require_relative 'expression/methods/strfregexp'
+require_relative 'expression/methods/tests'
+require_relative 'expression/methods/traverse'

data/lib/regexp_parser/lexer.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 # A very thin wrapper around the scanner that breaks quantified literal runs,
 # collects emitted tokens into an array, calculates their nesting depth, and
 # normalizes tokens for the parser, and checks if they are implemented by the

data/lib/regexp_parser/parser.rb

@@ -1,5 +1,7 @@
-require 'regexp_parser/error'
-require 'regexp_parser/expression'
+# frozen_string_literal: true
+
+require_relative 'error'
+require_relative 'expression'
 
 class Regexp::Parser
   include Regexp::Expression
@@ -319,6 +321,7 @@ class Regexp::Parser
     when :codepoint_list; node << EscapeSequence::CodepointList.new(token, active_opts)
     when :hex;            node << EscapeSequence::Hex.new(token, active_opts)
     when :octal;          node << EscapeSequence::Octal.new(token, active_opts)
+    when :utf8_hex;       node << EscapeSequence::UTF8Hex.new(token, active_opts)
 
     when :control
       if token.text =~ /\A(?:\\C-\\M|\\c\\M)/
@@ -467,6 +470,7 @@ class Regexp::Parser
     when *UPTokens::Age;          node << UP::Age.new(token, active_opts)
     when *UPTokens::Derived;      node << UP::Derived.new(token, active_opts)
     when *UPTokens::Emoji;        node << UP::Emoji.new(token, active_opts)
+    when *UPTokens::Enumerated;   node << UP::Enumerated.new(token, active_opts)
     when *UPTokens::Script;       node << UP::Script.new(token, active_opts)
     when *UPTokens::UnicodeBlock; node << UP::Block.new(token, active_opts)
 
@@ -574,21 +578,24 @@ class Regexp::Parser
     options_stack.last
   end
 
-  # Assigns referenced expressions to refering expressions, e.g. if there is
+  # Assigns referenced expressions to referring expressions, e.g. if there is
   # an instance of Backreference::Number, its #referenced_expression is set to
   # the instance of Group::Capture that it refers to via its number.
   def assign_referenced_expressions
-    # find all referencable and refering expressions
-    targets = { 0 => root }
+    # find all referenceable and referring expressions
+    targets = { 0 => [root] }
     referrers = []
     root.each_expression do |exp|
-      exp.is_a?(Group::Capture) && targets[exp.identifier] = exp
-      referrers << exp if exp.referential?
+      if exp.referential?
+        referrers << exp
+      elsif exp.is_a?(Group::Capture)
+        (targets[exp.identifier] ||= []) << exp
+      end
     end
-    # assign reference expression to refering expressions
+    # assign referenced expressions to referring expressions
     # (in a second iteration because there might be forward references)
     referrers.each do |exp|
-      exp.referenced_expression = targets[exp.reference] ||
+      exp.referenced_expressions = targets[exp.reference] ||
         raise(ParserError, "Invalid reference #{exp.reference} at pos #{exp.ts}")
     end
   end

data/lib/regexp_parser/scanner/errors/premature_end_error.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 class Regexp::Scanner
   # Unexpected end of pattern
   class PrematureEndError < ScannerError

data/lib/regexp_parser/scanner/errors/scanner_error.rb

@@ -1,4 +1,6 @@
-require 'regexp_parser/error'
+# frozen_string_literal: true
+
+require_relative '../../../regexp_parser/error'
 
 class Regexp::Scanner
   # General scanner error (catch all)

data/lib/regexp_parser/scanner/errors/validation_error.rb

@@ -1,3 +1,5 @@
+# frozen_string_literal: true
+
 class Regexp::Scanner
   # Base for all scanner validation errors
   class ValidationError < ScannerError

data/lib/regexp_parser/scanner/properties/long.csv

@@ -8,6 +8,8 @@ age=12.1,age=12.1
 age=13.0,age=13.0
 age=14.0,age=14.0
 age=15.0,age=15.0
+age=15.1,age=15.1
+age=16.0,age=16.0
 age=2.0,age=2.0
 age=2.1,age=2.1
 age=3.0,age=3.0
@@ -102,18 +104,33 @@ extendedpictographic,extended_pictographic
 extender,extender
 finalpunctuation,final_punctuation
 format,format
+garay,garay
 georgian,georgian
 glagolitic,glagolitic
 gothic,gothic
 grantha,grantha
 graph,graph
 graphemebase,grapheme_base
+graphemeclusterbreak=control,grapheme_cluster_break=control
+graphemeclusterbreak=cr,grapheme_cluster_break=cr
+graphemeclusterbreak=extend,grapheme_cluster_break=extend
+graphemeclusterbreak=l,grapheme_cluster_break=l
+graphemeclusterbreak=lf,grapheme_cluster_break=lf
+graphemeclusterbreak=lv,grapheme_cluster_break=lv
+graphemeclusterbreak=lvt,grapheme_cluster_break=lvt
+graphemeclusterbreak=prepend,grapheme_cluster_break=prepend
+graphemeclusterbreak=regionalindicator,grapheme_cluster_break=regional_indicator
+graphemeclusterbreak=spacingmark,grapheme_cluster_break=spacingmark
+graphemeclusterbreak=t,grapheme_cluster_break=t
+graphemeclusterbreak=v,grapheme_cluster_break=v
+graphemeclusterbreak=zwj,grapheme_cluster_break=zwj
 graphemeextend,grapheme_extend
 graphemelink,grapheme_link
 greek,greek
 gujarati,gujarati
 gunjalagondi,gunjala_gondi
 gurmukhi,gurmukhi
+gurungkhema,gurung_khema
 han,han
 hangul,hangul
 hanifirohingya,hanifi_rohingya
@@ -123,11 +140,14 @@ hebrew,hebrew
 hexdigit,hex_digit
 hiragana,hiragana
 hyphen,hyphen
+idcompatmathcontinue,id_compat_math_continue
+idcompatmathstart,id_compat_math_start
 idcontinue,id_continue
 ideographic,ideographic
 idsbinaryoperator,ids_binary_operator
 idstart,id_start
 idstrinaryoperator,ids_trinary_operator
+idsunaryoperator,ids_unary_operator
 imperialaramaic,imperial_aramaic
 inadlam,in_adlam
 inaegeannumbers,in_aegean_numbers
@@ -190,6 +210,7 @@ incjkunifiedideographsextensione,in_cjk_unified_ideographs_extension_e
 incjkunifiedideographsextensionf,in_cjk_unified_ideographs_extension_f
 incjkunifiedideographsextensiong,in_cjk_unified_ideographs_extension_g
 incjkunifiedideographsextensionh,in_cjk_unified_ideographs_extension_h
+incjkunifiedideographsextensioni,in_cjk_unified_ideographs_extension_i
 incombiningdiacriticalmarks,in_combining_diacritical_marks
 incombiningdiacriticalmarksextended,in_combining_diacritical_marks_extended
 incombiningdiacriticalmarksforsymbols,in_combining_diacritical_marks_for_symbols
@@ -223,6 +244,7 @@ induployan,in_duployan
 inearlydynasticcuneiform,in_early_dynastic_cuneiform
 inegyptianhieroglyphformatcontrols,in_egyptian_hieroglyph_format_controls
 inegyptianhieroglyphs,in_egyptian_hieroglyphs
+inegyptianhieroglyphsextendeda,in_egyptian_hieroglyphs_extended_a
 inelbasan,in_elbasan
 inelymaic,in_elymaic
 inemoticons,in_emoticons
@@ -235,6 +257,7 @@ inethiopicextended,in_ethiopic_extended
 inethiopicextendeda,in_ethiopic_extended_a
 inethiopicextendedb,in_ethiopic_extended_b
 inethiopicsupplement,in_ethiopic_supplement
+ingaray,in_garay
 ingeneralpunctuation,in_general_punctuation
 ingeometricshapes,in_geometric_shapes
 ingeometricshapesextended,in_geometric_shapes_extended
@@ -250,6 +273,7 @@ ingreekextended,in_greek_extended
 ingujarati,in_gujarati
 ingunjalagondi,in_gunjala_gondi
 ingurmukhi,in_gurmukhi
+ingurungkhema,in_gurung_khema
 inhalfwidthandfullwidthforms,in_halfwidth_and_fullwidth_forms
 inhangulcompatibilityjamo,in_hangul_compatibility_jamo
 inhanguljamo,in_hangul_jamo
@@ -291,6 +315,7 @@ inkhmer,in_khmer
 inkhmersymbols,in_khmer_symbols
 inkhojki,in_khojki
 inkhudawadi,in_khudawadi
+inkiratrai,in_kirat_rai
 inlao,in_lao
 inlatin1supplement,in_latin_1_supplement
 inlatinextendeda,in_latin_extended_a
@@ -346,6 +371,7 @@ inmusicalsymbols,in_musical_symbols
 inmyanmar,in_myanmar
 inmyanmarextendeda,in_myanmar_extended_a
 inmyanmarextendedb,in_myanmar_extended_b
+inmyanmarextendedc,in_myanmar_extended_c
 innabataean,in_nabataean
 innagmundari,in_nag_mundari
 innandinagari,in_nandinagari
@@ -367,6 +393,7 @@ inoldsogdian,in_old_sogdian
 inoldsoutharabian,in_old_south_arabian
 inoldturkic,in_old_turkic
 inolduyghur,in_old_uyghur
+inolonal,in_ol_onal
 inopticalcharacterrecognition,in_optical_character_recognition
 inoriya,in_oriya
 inornamentaldingbats,in_ornamental_dingbats
@@ -406,6 +433,7 @@ inspacingmodifierletters,in_spacing_modifier_letters
 inspecials,in_specials
 insundanese,in_sundanese
 insundanesesupplement,in_sundanese_supplement
+insunuwar,in_sunuwar
 insuperscriptsandsubscripts,in_superscripts_and_subscripts
 insupplementalarrowsa,in_supplemental_arrows_a
 insupplementalarrowsb,in_supplemental_arrows_b
@@ -419,6 +447,7 @@ insuttonsignwriting,in_sutton_signwriting
 insylotinagri,in_syloti_nagri
 insymbolsandpictographsextendeda,in_symbols_and_pictographs_extended_a
 insymbolsforlegacycomputing,in_symbols_for_legacy_computing
+insymbolsforlegacycomputingsupplement,in_symbols_for_legacy_computing_supplement
 insyriac,in_syriac
 insyriacsupplement,in_syriac_supplement
 intagalog,in_tagalog
@@ -441,8 +470,10 @@ inthai,in_thai
 intibetan,in_tibetan
 intifinagh,in_tifinagh
 intirhuta,in_tirhuta
+intodhri,in_todhri
 intoto,in_toto
 intransportandmapsymbols,in_transport_and_map_symbols
+intulutigalari,in_tulu_tigalari
 inugaritic,in_ugaritic
 inunifiedcanadianaboriginalsyllabics,in_unified_canadian_aboriginal_syllabics
 inunifiedcanadianaboriginalsyllabicsextended,in_unified_canadian_aboriginal_syllabics_extended
@@ -473,6 +504,7 @@ khitansmallscript,khitan_small_script
 khmer,khmer
 khojki,khojki
 khudawadi,khudawadi
+kiratrai,kirat_rai
 lao,lao
 latin,latin
 lepcha,lepcha
@@ -506,6 +538,7 @@ meroiticcursive,meroitic_cursive
 meroitichieroglyphs,meroitic_hieroglyphs
 miao,miao
 modi,modi
+modifiercombiningmark,modifier_combining_mark
 modifierletter,modifier_letter
 modifiersymbol,modifier_symbol
 mongolian,mongolian
@@ -535,6 +568,7 @@ oldsogdian,old_sogdian
 oldsoutharabian,old_south_arabian
 oldturkic,old_turkic
 olduyghur,old_uyghur
+olonal,ol_onal
 openpunctuation,open_punctuation
 oriya,oriya
 osage,osage
@@ -588,6 +622,7 @@ space,space
 spaceseparator,space_separator
 spacingmark,spacing_mark
 sundanese,sundanese
+sunuwar,sunuwar
 surrogate,surrogate
 sylotinagri,syloti_nagri
 symbol,symbol
@@ -609,7 +644,9 @@ tibetan,tibetan
 tifinagh,tifinagh
 tirhuta,tirhuta
 titlecaseletter,titlecase_letter
+todhri,todhri
 toto,toto
+tulutigalari,tulu_tigalari
 ugaritic,ugaritic
 unassigned,unassigned
 unifiedideograph,unified_ideograph

data/lib/regexp_parser/scanner/properties/short.csv

@@ -58,6 +58,7 @@ epres,emoji_presentation
 ethi,ethiopic
 ext,extender
 extpict,extended_pictographic
+gara,garay
 geor,georgian
 glag,glagolitic
 gong,gunjala_gondi
@@ -69,6 +70,7 @@ grek,greek
 grext,grapheme_extend
 grlink,grapheme_link
 gujr,gujarati
+gukh,gurung_khema
 guru,gurmukhi
 hang,hangul
 hani,han
@@ -86,6 +88,7 @@ ideo,ideographic
 ids,id_start
 idsb,ids_binary_operator
 idst,ids_trinary_operator
+idsu,ids_unary_operator
 ital,old_italic
 java,javanese
 joinc,join_control
@@ -96,6 +99,7 @@ khmr,khmer
 khoj,khojki
 kits,khitan_small_script
 knda,kannada
+krai,kirat_rai
 kthi,kaithi
 l,letter
 lana,tai_tham
@@ -121,6 +125,7 @@ mand,mandaic
 mani,manichaean
 marc,marchen
 mc,spacing_mark
+mcm,modifier_combining_mark
 me,enclosing_mark
 medf,medefaidrin
 mend,mende_kikakui
@@ -153,6 +158,7 @@ oids,other_id_start
 olck,ol_chiki
 olower,other_lowercase
 omath,other_math
+onao,ol_onal
 orkh,old_turkic
 orya,oriya
 osge,osage
@@ -207,6 +213,7 @@ sora,sora_sompeng
 soyo,soyombo
 sterm,sentence_terminal
 sund,sundanese
+sunu,sunuwar
 sylo,syloti_nagri
 syrc,syriac
 tagb,tagbanwa
@@ -224,6 +231,8 @@ thaa,thaana
 tibt,tibetan
 tirh,tirhuta
 tnsa,tangsa
+todr,todhri
+tutg,tulu_tigalari
 ugar,ugaritic
 uideo,unified_ideograph
 vaii,vai

data/lib/regexp_parser/scanner/scanner.rl

@@ -37,7 +37,8 @@
   octal_sequence        = [0-7]{1,3};
 
   hex_sequence          = 'x' . xdigit{1,2};
-  hex_sequence_err      = 'x' . [^0-9a-fA-F{];
+  hex_sequence_err      = 'x' . [^0-9A-Fa-f];
+  high_hex_sequence     = 'x' . [89A-Fa-f] . xdigit . ( '\\x' . [89A-Fa-f] . xdigit )*;
 
   codepoint_single      = 'u' . xdigit{4};
   codepoint_list        = 'u{' . xdigit{1,6} . (space . xdigit{1,6})* . '}';
@@ -78,8 +79,8 @@
   # try to treat every other group head as options group, like Ruby
   group_options         = '?' . ( [^!#'():<=>~]+ . ':'? ) ?;
 
-  group_name_id_ab      = ([^!0-9\->] | utf8_multibyte) . ([^>] | utf8_multibyte)*;
-  group_name_id_sq      = ([^0-9\-']  | utf8_multibyte) . ([^'] | utf8_multibyte)*;
+  group_name_id_ab      = ([^!=0-9\->] | utf8_multibyte) . ([^>] | utf8_multibyte)*;
+  group_name_id_sq      = ([^0-9\-']   | utf8_multibyte) . ([^'] | utf8_multibyte)*;
   group_number          = '-'? . [0-9]+;
   group_level           = [+\-] . [0-9]+;
 
@@ -210,7 +211,7 @@
         type = :nonposixclass
       end
 
-      unless self.class.posix_classes.include?(class_name)
+      unless POSIX_CLASSES[class_name]
         raise ValidationError.for(:posix_class, text)
       end
 
@@ -246,7 +247,7 @@
     # Treat all remaining escapes - those not supported in sets - as literal.
     # (This currently includes \^, \-, \&, \:, although these could potentially
     # be meta chars when not escaped, depending on their position in the set.)
-    any > (escaped_set_alpha, 1) {
+    (any | utf8_multibyte) > (escaped_set_alpha, 1) {
       emit(:escape, :literal, copy(data, ts-1, te))
       fret;
     };
@@ -256,9 +257,21 @@
   # escape sequence scanner
   # --------------------------------------------------------------------------
   escape_sequence := |*
-    [1-9] {
+    [1-9] . [0-9]* {
       text = copy(data, ts-1, te)
-      emit(:backref, :number, text)
+
+      # If not enough groups have been opened, there is a fallback to either an
+      # octal or literal interpretation for 2+ digit numerical escapes.
+      digits = text[1..-1]
+      if digits.size == 1 || digits.to_i <= capturing_group_count
+        emit(:backref, :number, text)
+      elsif digits =~ /\A[0-7]{2,}\z/
+        emit(:escape, :octal, text)
+      else
+        emit(:escape, :literal, text[0..1])
+        emit(:literal, :literal, text[2..-1])
+      end
+
       fret;
     };
 
@@ -267,6 +280,13 @@
       fret;
     };
 
+    [8-9] . [0-9] { # special case, emits two tokens
+      text = copy(data, ts-1, te)
+      emit(:escape, :literal, text[0, 2])
+      emit(:literal, :literal, text[2])
+      fret;
+    };
+
     meta_char {
       case text = copy(data, ts-1, te)
       when '\.';  emit(:escape, :dot,               text)
@@ -314,6 +334,16 @@
       fret;
     };
 
+    high_hex_sequence > (escaped_alpha, 5) {
+      text = copy(data, ts-1, te)
+      if regexp_encoding == Encoding::BINARY
+        text.split(/(?=\\)/).each { |part| emit(:escape, :hex, part) }
+      else
+        emit(:escape, :utf8_hex, text)
+      end
+      fret;
+    };
+
     hex_sequence > (escaped_alpha, 5) @eof(premature_end_error) {
       emit(:escape, :hex, copy(data, ts-1, te))
       fret;
@@ -357,6 +387,7 @@
   conditional_expression := |*
     group_lookup . ')' {
       text = copy(data, ts, te-1)
+      text =~ /[^0]/ or raise ValidationError.for(:backref, 'condition', 'invalid ref ID')
       emit(:conditional, :condition, text)
       emit(:conditional, :condition_close, ')')
     };
@@ -506,6 +537,7 @@
     };
 
     group_open @group_opened {
+      self.capturing_group_count = capturing_group_count + 1
       text = copy(data, ts, te)
       emit(:group, :capture, text)
     };
@@ -534,13 +566,13 @@
       case text = copy(data, ts, te)
       when /^\\k(.)[^0-9\-][^+\-]*['>]$/
         emit(:backref, $1 == '<' ? :name_ref_ab : :name_ref_sq, text)
-      when /^\\k(.)[1-9]\d*['>]$/
+      when /^\\k(.)0*[1-9]\d*['>]$/
         emit(:backref, $1 == '<' ? :number_ref_ab : :number_ref_sq, text)
-      when /^\\k(.)-[1-9]\d*['>]$/
+      when /^\\k(.)-0*[1-9]\d*['>]$/
         emit(:backref, $1 == '<' ? :number_rel_ref_ab : :number_rel_ref_sq, text)
       when /^\\k(.)[^0-9\-].*[+\-]\d+['>]$/
         emit(:backref, $1 == '<' ? :name_recursion_ref_ab : :name_recursion_ref_sq, text)
-      when /^\\k(.)-?[1-9]\d*[+\-]\d+['>]$/
+      when /^\\k(.)-?0*[1-9]\d*[+\-]\d+['>]$/
         emit(:backref, $1 == '<' ? :number_recursion_ref_ab : :number_recursion_ref_sq, text)
       else
         raise ValidationError.for(:backref, 'backreference', 'invalid ref ID')
@@ -553,9 +585,9 @@
       case text = copy(data, ts, te)
       when /^\\g(.)[^0-9+\-].*['>]$/
         emit(:backref, $1 == '<' ? :name_call_ab : :name_call_sq, text)
-      when /^\\g(.)\d+['>]$/
+      when /^\\g(.)(?:0|0*[1-9]\d*)['>]$/
         emit(:backref, $1 == '<' ? :number_call_ab : :number_call_sq, text)
-      when /^\\g(.)[+-]\d+/
+      when /^\\g(.)[+-]0*[1-9]\d*/
         emit(:backref, $1 == '<' ? :number_rel_call_ab : :number_rel_call_sq, text)
       else
         raise ValidationError.for(:backref, 'subexpression call', 'invalid ref ID')
@@ -632,9 +664,9 @@
   *|;
 }%%
 
-require 'regexp_parser/scanner/errors/scanner_error'
-require 'regexp_parser/scanner/errors/premature_end_error'
-require 'regexp_parser/scanner/errors/validation_error'
+require_relative 'scanner/errors/scanner_error'
+require_relative 'scanner/errors/premature_end_error'
+require_relative 'scanner/errors/validation_error'
 
 class Regexp::Scanner
   # Scans the given regular expression text, or Regexp object and collects the
@@ -654,6 +686,7 @@ class Regexp::Scanner
 
     input = input_object.is_a?(Regexp) ? input_object.source : input_object
     self.free_spacing = free_spacing?(input_object, options)
+    self.regexp_encoding = extract_encoding(input_object, options)
     self.spacing_stack = [{:free_spacing => free_spacing, :depth => 0}]
 
     data  = input.unpack("c*")
@@ -664,6 +697,7 @@ class Regexp::Scanner
 
     self.set_depth = 0
     self.group_depth = 0
+    self.capturing_group_count = 0
     self.conditional_stack = []
     self.char_pos = 0
 
@@ -703,10 +737,11 @@ class Regexp::Scanner
     File.read("#{__dir__}/scanner/properties/#{name}.csv").scan(/(.+),(.+)/).to_h
   end
 
-  def self.posix_classes
+  # Use each_with_object for required_ruby_version >= 2.2, or #to_h for >= 2.6
+  POSIX_CLASSES =
     %w[alnum alpha ascii blank cntrl digit graph
        lower print punct space upper word xdigit]
-  end
+      .inject({}) { |o, e| o.merge(e => true) }.freeze
 
   # Emits an array with the details of the scanned pattern
   def emit(type, token, text)
@@ -734,16 +769,25 @@ class Regexp::Scanner
     end
   end
 
-  attr_accessor :literal_run # only public for #||= to work on ruby <= 2.5
+  attr_accessor :capturing_group_count, :literal_run # only public for #||= to work on ruby <= 2.5
 
   private
 
   attr_accessor :block,
                 :collect_tokens, :tokens, :prev_token,
                 :free_spacing, :spacing_stack,
+                :regexp_encoding,
                 :group_depth, :set_depth, :conditional_stack,
                 :char_pos
 
+  def extract_encoding(input_object, options)
+    if input_object.is_a?(::Regexp)
+      input_object.encoding
+    elsif options && (options & Regexp::NOENCODING)
+      Encoding::BINARY
+    end
+  end
+
   def free_spacing?(input_object, options)
     if options && !input_object.is_a?(String)
       raise ArgumentError, 'options cannot be supplied unless scanning a String'